Final Report
PROVISIONAL ALGAL
ASSAY PROCEDURES
SANITARY ENGINEERING RESEARCH LABORATORY
COLLEGE OF ENGINEERING
AND
SCHOOL OF PUBLIC HEALTH
UNIVERSITY OF CALIFORNIA
BERKELEY
SERL REPORT No. 71-6
-------
WATER POLLUTION CONTROL RESEARCH SERIES
The Water Pollution Control Research Series describes
the results and progress .in the control and abatement
of pollution in our Nation's waters. They provide a
central source of information on the research, develop-
ment, and demonstration activities in the Water Quality
Office, Environmental Protection Agency, through inhouse
research and grants and contracts with Federal, State,
and .local agencies, research institutions, and industrial
organizations.
Inquiries pertaining to Water Pollution Control Research
Reports should be directed to the Head, Project Reports
System, Office of Research and Development, Water Quality
Office, Environmental Protection Agency, Room 1108,
Washington, D. C. 20242.
-------
FINAL REPORT
PROVISIONAL ALGAL ASSAY PROCEDURES
By
D. F. Toerien
C. H, Huang
J. Radimsky
E. A. Pearson
J, Scherfig
for the
WATER QUALITY OFFICE
ENVIRONMENTAL PROTECTION AGENCY
Project No. 16010 DQB
October 1971
Sanitary Engineering Research Laboratory
College of Engineering
and
School of Public Health
University of California
Berkeley
SERL Report No. 71-6
-------
Notice
The contents of this report do not necessarily
reflect the views and policies of the Environ-
mental Protection Agency, nor does mention
of trade names or commercial products con-
stitute endorsement or recommendation for
use.
ii
-------
ABSTRACT
Batch and continuous flow (chemostat) assays were investigated as
part of a joint industry-government sponsored, multilaboratory
effort to develop a standardized algal assay procedure for nutrient
level assessment. Assays were conducted with Selenastrum capri-
cornutum as a standard assay organism.
Batch culture assays were found to have a lower level of precision
than chemostat assays in the assessment of growth response as a
function of nutrient concentration. The biomass parameter, maximum
cell concentration, X, of the batch assay generally responded to the
nutrient concentration of the sample; however, the chemostat biomass
parameter, steady state cell concentration, Xp always was found to
be proportional to the nutrient concentration of the samples.
The results of spiking tests with batch assays generally were incon-
clusive with respect to identification of the growth rate limiting nutrient
whereas the results of spiking tests with chemostats indicated clearly
the growth rate limiting nutrient. It is recommended that batch type
algal assays be used only for crude screening or routine monitoring
purposes and that the chemostat should be used for the quantitative
assessment of the algal growth supporting properties of waters as
well as for the development of kinetic descriptions for nuisance algae
and the rate limiting nutrients of concern.
A kinetic description of Selenastrum capricornutum indicated a low
half saturation constant, Ks, (the concentration of nutrient supporting
one-half the maximum growth rate) of about 5 |j.g P/? for phosphate
phosphorus and a yield coefficient, Y, that varied as a function of
growth rate. A theoretical model was proposed and evaluated which
describes the varying yield coefficient (the result of "excess" uptake)
as a function of the growth rate (mean cell residence time). The
function was verified experimentally at a very high statistical confidence
level. The significance of these findings and their application to the
practical problem of eutrophication assessment is presented.
This report is submitted in accordance with the requirements of Grant
No, 16010 DQB of the Water Quality Office, Environmental Protection
Agency.
iii
-------
CONTENTS
Section Page
I Conclusions 1
II Recommendations 11
III Introduction 15
IV Theory and Rationale 25
V Materials and Methods 59
VI Results and Discussion 71
VII Comparison of Batch and Chemostat Algal Assays 161
VIII Acknowledgments 167
IX References 169
X Publications and Patents 179
XI Glossary 181
XII Appendices 183
v
-------
1
2
3
4
5
6
7
8
9
10
1 i
12
FIGURES
PAGE
ALGAL GROWTH MODEL FOR NATURAL AQUATIC
SYSTEMS 16
A GRAPHICAL COMPARISON OF MIC HA ELIS - MEN TEN,
MONOD EQUATION, EXPONENTIAL EQUATION AND
BLACKMAN MODEL WITH THE SAME VALUES OF
MAXIMUM SPECIFIC GROWTH RATE AND HALF
SATURATION CONSTANT 29
IDEAL BATCH GROWTH CURVE 33
BATCH ALGAL GROWTH IN UNRESTRICTED
NUTRIENT ENVIRONMENTS 35
BATCH ALGAL GROWTH UNDER SPECIFIC GROWTH
RATE LIMITING CONDITIONS 36
CONTINUOUS FLOW STIRRED TANK REACTOR
SYSTEM (CFSTR) 40
THEORETICAL STEADY STATE CONDITIONS IN
CHEMOSTAT 43
COMPARISON OF NUTRIENT REMOVAL VELOCITY
AND NUTRIENT UTILIZATION RATE OF NO "EXCESS"
AND "EXCESS" NUTRIENT UPTAKE CONDITIONS 46
DEFINITION SKETCH FOR A SPECIFIC NUTRIENT
SUPPLY, UPTAKE AND UTILIZATION FOR GROWTH
IN A CHEMOSTAT AT STEADY STATE 48
COMPONENT NUTRIENT CONCENTRATIONS FOR
VARYING MEAN CELL AGES FOR A SPECIFIC
NUTRIENT IN STEADY STATE CHEMOSTAT OPERA-
TION WHEN "EXCESS" NUTRIENT UPTAKE OCCURS 49
GRAPHICAL COMPARISON OF EXPONENTIAL
FUNCTION AND RECTANGULAR HYPERBOLIC FUNCTION
FOR NET YIELD COEFFICIENT - CELL AGE RELATION-
SHIP (SAME VALUE OF MAXIMUM CELL YIELD)
ymax} 52
THE FRACTION OF EXCESS NUTRIENT UPTAKE AS
A FUNCTION OF MEAN CELL AGE (COMPARISON
OF EXPONENTIAL AND HYPERBOLIC MODELS) 54
VI
-------
PAGE
13 THEORECTICAL STEADY STATE CONDITIONS
IN CHEMOSTAT WITH A VARYING YIELD
COEFFICIENT (EXCESS NUTRIENT UPTAKE)
14 AIR TREATMENT AND DISTRIBUTION SYSTEM
15 BATCH ASSAY UNIT
16 DIAGRAM OF CHEMOSTAT ASSAY UNIT
17 FLOW DIAGRAM OF ROUTINE ANALYSES
PERFORMED ON CHEMOSTATS
18 RELATIONSHIP OF SELENASTRUM CAPRICORNUTUM
RESPONSE (SUSPENDED SOLIDS) TO CONCENTRATION
OF PAAP MEDIUM
19 RELATIONSHIP OF SELENASTRUM CAPRICORNUTUM
RESPONSE (CELL NUMBERS) TO CONCENTRATION OF
PAAP MEDIUM
20 RELATIONSHIP OF SELENASTRUM CAPRICORNUTUM
RESPONSE (TOTAL CELL VOLUME) TO CONCENTRA-
TION OF PAAP MEDIUM
21 RELATIONSHIP BETWEEN DILUTION OF CLEAR
LAKE WATER AND X BASED ON SUSPENDED
SOLIDS
22 RELATIONSHIP BETWEEN DILUTION OF CLEAR
LAKE WATER AND X BASED ON CELL NUMBERS
23 RELATIONSHIP BETWEEN DILUTION OF CLEAR
LAKE WATER AND X BASED ON TOTAL CELL
VOLUME
24 MAXIMUM BATCH GROWTH RATES (jl. ) IN
CLEAR LAKE WATER 0
25 EFFECT OF SECONDARY EFFLUENT ADDITION
ON X BASED ON SUSPENDED SOLIDS OF CLEAR
LAKE WATER
26 EFFECT OF SECONDARY EFFLUENT ADDITION ON
X BASED ON CELL NUMBERS OF CLEAR LAKE
WATER
57
66
67
68
70
81
82
84
91
91
92
94
99
99
vii
-------
27
28
29
30
31
32
33
34
35
36
37
38
39
100
102
104
107
107
108
110
116
116
117
117
119
121
EFFECT OF SECONDARY EFFLUENT ADDITION
ON X BASED ON TOTAL CELL VOLUME OF
CLEAR LAKE WATER
EFFECT OF SECONDARY EFFLUENT ON THE
ALGAL GROWTH RATE OF CLEAR LAKE
WATER
EFFECT OF ENRICHMENT (SPIKING) ON X
OF CLEAR LAKE WATER
RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND X BASED
ON SUSPENDED SOLIDS
RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND X BASED
ON CELL NUMBERS
RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND X BASED
ON TOTAL CELL VOLUME
MAXIMUM BATCH GROWTH RATE (pu ) IN
SACRAMENTO RIVER WATER
EFFECT OF SECONDARY EFFLUENT ADDITION
ON X BASED ON CELL NUMBERS OF SACRAMENTO
RIVER WATER
EFFECT OF SECONDARY EFFLUENT ADDITION
ON X BASED ON TOTAL CELL VOLUME OF
SACRAMENTO RIVER WATER
EFFECT OF SECONDARY EFFLUENT ADDITION
ON X BASED ON SUSPENDED SOLIDS OF
SACRAMENTO RIVER WATER
EFFECT OF SECONDARY EFFLUENT ADDITION
ON ^ OF SACRAMENTO RIVER WATER
EFFECT OF ENRICHMENT (SPIKING) ON THE X
OF SACRAMENTO RIVER WATER
GROWTH CURVES OF ENRICHMENTS (SPIKES)
OF SACRAMENTO RIVER WATER
viii
-------
PAGE
40 RELATIONSHIP BETWEEN CELL BIOMASS AND
PHOSPHORUS UPTAKE FOR DIFFERENT
RESIDENCE TIMES 126
41
OPERATION OF
CHEMOSTATS
1,
2,'AND 3
127
42
OPERATION OF
CHEMOSTATS
4,
5, AND 6
128
43
OPERATION OF
CHEMOSTATS
7,
8, AND 9
129
44
OPERATION OF
CHEMOSTATS
10,
11, AND 12
130
45 RELATIONSHIP BETWEEN STEADY STATE CELL
CONCENTRATION AND PHOSPHATE UPTAKE AT
INFINITE HYDRAULIC RESIDENCE TIME 140
46 NET CELL YIELD COEFFICIENT AS A FUNCTION
OF MEAN CELL AGE, 0c 141
47 COMPARISON OF NUTRIENT REMOVAL VELOCITY,
q, AND SPECIFIC GROWTH RATE, n 143
48 FRACTION OF "EXCESS" PHOSPHORUS UPTAKE
AS FUNCTION OF MEAN CELL AGE, 6 145
c
49 PLOT OF MICHAELIS-MENTEN (MONOD)
EQUATION FOR PHOSPHORUS LIMITATION 147
50 RELATIONSHIP BETWEEN STEADY STATE CELL
CONCENTRATION AND DILUTION OF
SACRAMENTO RIVER WATER 153
51 SACRAMENTO RIVER CHEMOSTAT CELL
CONCENTRATION, Xv VARIATION WITH
DIFFERENT SPIKING TREATMENTS 159
IX
-------
TABLES
No. Page
I Macronutrient Composition of 100% PAAP
Medium 60
II Micronutrient Composition of 100% PAAP
Medium 60
III Accuracy and Precision of Analytical
Methods and Biomass Estimates 64
IV Removal of Algae from PAAP Medium by
Pasteurization and Filtration 72
V Removal of Indigenous Algae from a Natural
Sample by Pretreatments 73
VI Pretreatment Effect on the Chemical
Composition of Samples 74
VII Mean Algal Growth Rates Obtained After
Sample Pretreatment 76
VIII Mean Cell Production (ft) Values Obtained
After Sample Pretreatment 77
IX Mean Values and Precision Estimates for
ft Based on Different Biomass Parameters 85
X Experimental Procedure for Identification of
Growth-Limiting Nutrient 89
XI Means and Precision of Maximum Cell
Concentration Estimates in Clear Lake Water 93
XII Means and Precision of Algal Growth Rate
Estimates for Clear Lake Water 95
XIII Means of Inorganic Nitrogen and Phosphorus
Concentrations in Clear Lake Water 96
XIV Effect of Nutrient Enrichment (Spikes) on
Mean Specific Growth Rate, Batch flx^) of
Clear Lake Water 106
XV Algal Growth Rate Estimates, for
Sacramento River Water 111
x
-------
Page
XVI Inorganic Nitrogen and Phosphorus
Concentrations in Sacramento River
Water
XVII Effect of Nutrient Enrichment (Spikes) on
Mean Specific Growth Rate, Batch (pL^) of
Sacramento River Water
XVIII Conditions for Chemostat Assays
XIX Precision of Chemostat Bioassay Procedure
XX Summary of Chemostat Runs A and B
XXI Summary of Chemostat Runs C and D
XXII Summary of Chemostat Runs E and F
XXIII Summary of Chemostat Runs G and H
XXIV Summary of Chemostat Runs I and J
XXV Kinetic Growth Models and Constants for
Selenastrum capricornutum Under
Phosphorus Limiting Conditions (Various
Models and Computational Methods)
XXVI The Most Probable Kinetic Growth Models
and Best Estimate of Growth Constants and
Coefficients for Selenastrum capricornutum
Under Phosphorus Limited Conditions
XXVII Summary of Chemostat Operation Results on
Sacramento River Water
XXVIII Presumed Steady State Chemostat Parameters
for Sacramento River Water
XXIX Chemostat Enrichment (Spiking) Procedure
and Observed Data with Sacramento River
Water
112
120
125
132
133
134
135
136
137
149
150
152
155
158
XI
-------
SECTION I
CONCLUSIONS
SAMPLE PRETREATMENT
1. Pasteurization and Whatman No. 41 and membrane (0. 45-jo. pore-
size) filtration removed test algae from simulated natural waters
based upon the absence of algal growth in treated samples.
2. Heat treatment, i. e. , pasteurization and autoclaving, and mem-
brane filtration removed indigenous algae from natural water sample.
However, Whatman No. 41 filtration did not appear to remove the
indigenous algae.
3. The chemical composition of samples apparently is not substantially
changed by heat treatment; however, filtration of samples removed
particulate matter from the samples which may reduce available
nutrient content.
4. The reduction of nutrient content by filtration techniques (especially
membrane filtration) resulted in significantly lower maximum cell con-
centrations (X) and maximum specific growth rates, batch (pt^) (p =
0. 95) in the filtered samples as compared to heat treated samples.
5. The removal of particulate matter from a sample appeared to
reduce the overall available nutrient concentrations of the sample,
thereby causing an underestimate of the X of that sample. The release
of previously bound nutrients (in particulate matter) by heat treatment
of samples increased the significantly, In the original untreated
water these nutrients were not immediately available for algal growth
and hence the (1^ was overestimated.
INTERLABORATORY PRECISION TEST
6. The use of foam rubber stoppers in batch assays can lead to toxic
conditions which affect algal growth. This toxicity appeared to persist
even after thorough washing procedures and caused growth inhibition
in the same flasks during subsequent assays.
7. To determine whether or not algal growth responses in assays were
affected by foam rubber toxicity, the foam rubber stoppers require
cumbersome evaluation which reduces the suitability of such stoppers
for batch assays.
8. Nontoxic stoppers, or glass beakers, etc., should be used in batch
assays rather than foam rubber stoppers.
1
-------
9. The X in full strength (100%) PAAP medium appeared to be
limited by something other than dissolved nutrients in the medium
because a proportional increase of with increase in nutrient con-
centration was not observed.
10. The growth limitation likely was caused by a limitation in the
availability of carbon (COz ) in the high nutrient concentration samples,
as evidenced by the high pH values reached in these assays.
11. Carbon limitation apparently can occur in high nutrient content
samples even when aeration is practiced. The use of specific surface
area to volume ratio flasks is no guarantee that carbon limitation will
not occur in high nutrient content samples. Even when aeration is
used, the pH of cultures must be kept between 7. 0 and 8. 3 by increas-
ing or decreasing air flow rates (or addition of COz ) to prevent carbon
from becoming limiting or the samples from having too low a pH.
12. The X responses for enriched samples (PAAP medium with initial
concentrations of more than 62 jig P/d and 1400 (ig N/0 as measured in
terms of suspended solids (mg/£), cell numbers (cells x 106/m2) or
total cell volume (jx3 x 106/m?) were similar, indicating that all three
methods were adequate parameters of biomass.
A
13. The adequacy of X responses for assays with low concentrations
of nutrients (i. e. , « 62 (xg P/i and « 1400 |ag NIt) has yet to be
demonstrated,
14. The fact that there was virtually no difference between the
values in die different concentrations of the PAAP medium tested
(i. e. , 10%, 30%, and 100%) is explained by the fact that the nutrient
level of the medium was so high, at least more than 10 times the Kfl
(half saturation constant) value for Selenastrum capricornutum, that
no differences should be expected. It does not indicate that is a
useless algal response.
15. The precision of X and jx, estimates in the enriched media assays
were adequate, i. e. , they had coefficients of variations of 15% or less;
values which are acceptable for biological assay techniques.
16. The results of the interlaboratory precision test suggest that the
batch bioassay method can be used with reasonably enriched samples
in eutrophication analyses, at least as a screening procedure, because
and can be estimated fairly precisely (coefficient of variation
approximately 15%),
NATURAL WATER SAMPLES
17. The & r esponse with the batch culture assay was reasonably
proportional to the concentration of the water samples as illustrated
2
-------
A
by the linear relationships between X responses and the concentration
(30% or more) of either Clear Lake or the Sacramento River.
A
18. The linear relationships between X and concentrations of the
natural waters illustrated the absence of toxicity in both waters.
19. The precision (coefficient of variation ^ 18. 5%) of X responses in
the natural samples indicated that X can be estimated fairly precisely
in natural water samples (e, g. , Sacramento River) having nutrient
concentrations in the order of 34 (jig Pit and 133 (ig NIt. The precision
of jxL responses was adequate (coefficient of variation 2 15%) except in
one triplicate set of flasks.
20. Clear Lake (a eutrophic lake with 387 |ig P/2 and 190 (JLg N/£) water
must be diluted more than 50% before the (1^ is appreciably reduced;
however, dilution of Sacramento River water reduced the even at one-
third dilution.
21. The low initial inorganic N/P ratios of Clear Lake (0. 49) and
Sacramento River (3. 9) indicated that nitrogen rather than phosphorus
was the growth-limiting nutrient; this was borne out by the yield coef-
ficients for phosphorus (Yp) in the assays. However, the yield
coefficients for nitrogen (Yj^) did not indicate that nitrogen limited
growth in the samples, possibly because hydrolysis of nitrogenous
particulate matter contributed nitrogen for cell synthesis.
22. The use of yield coefficients in batch assays for identifying the
growth-limiting nutrient is limited both by the release of nutrients
from organic particulate matter as well as by variations of yield
coefficients in batch culture under different culture conditions.
23. The addition of secondary effluent up to 30% (V/V) increased the
X responses of Clear Lake and Sacramento River linearly but esti-
mates showed toxic effects of such additions at the 20% and 30% levels
in the case of Clear Lake. The (x^ of Sacramento River increased with
increased concentration of secondary effluent.
24. The solution of eutrophication problems depends heavily on the
identification of the growth-limiting nutrients in receiving waters and
wastewater—receiving water mixtures because only then can one be
certain as to which nutrient, if any, must be removed or reduced in
concentration. The amount of reduction needed can only be determined
from the growth kinetics of algae, i. e. , knowledge of the maximum
specific growth rates (fi), half saturation constants (Ks), yield coef-
ficients (Y), decay rates (k^), and other growth parameters for the
algae of concern.
25. The 1^ measurements in wastewater—receiving water^ assays indi-
cated the presence of toxicity which was not indicated by X measurements
in this investigation.
3
-------
26. Enrichment (spiking) techniques to identify the growth-limiting
nutrient in natural waters may result in anomalous results based on
X or estimates. Because micronutrients apparently were important
as growth-limiting nutrients in the waters studied, the amount of con-
tamination in laboratory macronutrient chemicals could have caused
these anomalies. Spiking procedures should be evaluated more
thoroughly than was possible in this investigation,
27. The present status of the batch assay is that of a screening device,
and possibly as a routine monitoring method. However, its direct
application to the solution of practical problems within a rational frame-
work is not yet apparent.
28. Further research on the batch assay is needed to establish its
value for determining algal growth kinetics, and to provide a rational
framework for applying the results to control the algal standing crop
in the receiving waters.
CONTINUOUS CULTURE STUDIES
29. Phosphorus is the sole rate-limiting nutrient when N/P ratios are
greater than 40 and when all other macronutrients and micronutrients
are in excess.
30. "Excess" phosphorus uptake does occur for Selenastrum capri-
cornutum in the continuous culture reactor (chemostat). Theoretical
relationships for "excess" nutrient uptake were developed and evaluated
experimentally with the following results:
a. Net cell yield (Y ) - cell age (6C) relationship
where Y = maximum yield coefficient, K& = nutrient (phosphorus)
X)T1'3L.2C J\
utilization constant.
b. Fraction of excess nutrient (phosphorus) uptake (f ) - cell age
(0c) relationship
S -0 / K. -6 /4„ 78
c e c A c
e = S-TS7 = e
o 1
where S = amount of "excess" nutrient (phosphorus) uptake.
4
-------
31. The maximum specific growth rate, is related to "washout"
residence time 0W as follows:
f = F" + kd
W
where is the decay constant of the organism, time*1. For
Selenastrum capricornutum in continuous cultures k
-------
Kinetic Growth Relationship
Net yield coefficient—cell age
relationship
(Modified Cell Continuity
Equation)
Best Estimated Value and Equations
Xn
Y =
LL
- c c = Y 1 - e
n S - S: max
¦e /k.
c A
v - oflr cell produced
max ~ mg P utilized
= 4. 78 days
Fractions of excess phosphorus
uptake—cell age relationship
f = e
e
-e Ik.
c A
= 4, 78 days
Specific growth rate—rate -
limiting nutrient concentration
relationship
Michaelis-Menton. JVIonod equation
^ k8 + s,
kd = 0. 001 day"1
jl = 1. 85 day-1
K = 3. 7 ~ 5, 7 (a,g/f as P
s
36, The Ks for phosphorus-limited conditions was found to be quite
low (Ks = 3. 7 ~ 5, 7 fxg P/|). This indicates that the low phosphate
concentrations found in many natural waters (<30 p.g P/.?) would
support the growth of Selenastrum capricornutum at rates between
0. 9 and ~ 1. 8 day"1. Note that jl was found to be i.85 day"1. The
low value obtained for Ks for phosphorus and Selenastrum capri-
cornutum indicates that many waters with high phosphorus concen-
trations (i. e. , ^50 fxg P/l) have phosphorus in considerable excess
and are not likely limited by phosphorus for organisms with charac-
teristics similar to Selenastrum capricornutum.
37. The kinetics of algal growth response to phosphorus-limited
condition can be determined easily with continuous flow algal chemo-
stats. Using appropriate kinetic models, it is possible to evaluate
the kinetic models and estimate the kinetic constants and coefficients
for algal systems (organisms and rate-limiting nutrients) and the
results can be applied and interpreted directly to practical eutrophica-
tion problems.
6
-------
38. The half saturation constant, Ks, along with the maximum specific
growth rate, ft, for particular algal species and nutrients are the most
important parameters in algal growth kinetics and their practical
application. The chemostat assay is the only procedure currently
known and evaluated that will yield the required kinetic information
for the algal species and nutrients of concern.
CHEMOSTAT STUDIES - SACRAMENTO RIVER WATER
39. A linear relationship between the steady state cell concentration,
Xj , and the concentration of Sacramento River water illustrated the
practical applicability of the chemostat assay procedure in studies of
relative algal response and toxicity evaluation. For Sacramento
River water, the linearity of steady state cell concentration and the
dilution of the river water was:
Cell counts (x 556 cells/m£) = 10. 8 + 1. 34% Sacramento
River water with a correlation coefficient of r = 0. 97 and
n = 9. No toxic effect on growth was observed.
40. The use of the yield coefficient in identifying the growth rate
limiting nutrient was limited. However, some positive conclusions
could be made from the yield coefficients with respect to certain
nutrients. For example, with Sacramento River water, nitrogen was
not the limiting nutrient because the nitrogen yield coefficients were
markedly lower than those that have been observed for nitrogen-
limited systems.
41. It appears that the most useful growth constants is the half
saturation constant, Ks. It can be used not only in identifying the
growth-limiting nutrient but also in the practical application of kinetic
data to real problems. Based upon the values obtained for Ks for
phosphorus, it can be positively concluded that phosphorus was not the
growth-limiting nutrient in Sacramento River water. However, no
definite conclusion can be made with respect to nitrogen based on Ks
values due to the lack of a reliable description of Selenastrum capri-
cornutum with respect to nitrogen-limited conditions.
42. A chemostat enrichment (spiking) procedure was used to identify
tiie growth-limiting nutrient in Sacramento River water. Nitrogen,
phosphorus, or iron were shown not to limit growth, whereas one of
the micronutrients in a micronutrient mix appeared to limit the growth
of Selenastrum capricornutum in Sacramento River water. The chemo-
stat spiking procedure appeared to be the easiest, quickest, and most
reliable of all spiking methods tested for the identification of the
growth-limiting nutrient.
7
-------
COMPARISON OF BATCH AND CHEMOSTAT ASSAYS
43. The biomass response of both the batch and the chemostat assays
was linearly related to the concentration of the growth-limiting
nutrient. The chemostat, however, had the highest precision.
44. The chemostat is more suitable than the batch assay for deter-
mining the growth rate—nutrient concentration relationships of
Selenastrum capricornutum.
45. The use of yield coefficients appeared to be of no value in identi-
fying the growth rate limiting nutrient in batch assays; however, with
chemostats it appeared to have promise once the relationships between
the yield coefficient with "excess" uptake and the growth rate have
been established as was done for phosphorus in this study.
46. Enrichment (spiking) techniques employed in this study were of
questionable value in batch assays when used to identify the growth-
limiting nutrient; however, with the chemostats they appeared to
identify the growth-limiting nutrient in question.
47. The Kg determinations are without value in batch assays for
identifying the growth-limiting nutrient; however, in chemostats they
can be extremely helpful,
48. Batch culture assays were markedly inferior to chemostats for
determining the growth kinetics of algae.
49. Batch culture assays have less equipment requirements than
chemostat assays and are simpler to conduct. However, chemostat
assays are more versatile and yield more valuable information with
higher precision than batch culture assays. This is especially true
for the important kinetic relationships,
50. The operational costs and time requirements of batch culture
assays and chemostat assays were not significantly different.
ROLE OF BATCH CULTURE ASSAYS IN
EUTROPHICATION ANALYSES
51, Batch culture assays, being inferior to chemostat assays in most
respects, appear to be best suited for screening and routine monitoring
purposes.
8
-------
ROLE OF CHEMOSTAT ASSAYS IN
EU TROPHIC A TION ANALYSES
52. Chemostat assays have a multiple role in eutrophication analyses.
a. The primary role of chemostats is for determination of the
kinetic characteristics (jx, Ks, Yn, Ymax> k
-------
SECTION II
RECOMMENDATIONS
SAMPLE PRETREATMENT
1. Filtration through Whatman No, 41 filter paper should not be used
as a sample pretreatment method because its ability to remove
indigenous algae is questionable.
2. The use of electronic particle counter algal counts for heat-treated,
nonfiltered samples must be made with caution because of the possibility
of artifact counts due to debris which contributes to the total count.
3. Heat treatment of samples, i. e. , pasteurization and autoclaving
(without a^subsequent filtration step), is recommended for samples
when the X is to be estimated. Membrane filtration (0. 45-fx pore size)
of the samples is recommended when is to be estimated. Of the
two heat treatments evaluated, autoclaving is recommended because
of its simplicity.
INTER LA BORA TORY PRECISION TEST
4. The use of foam rubber stoppers in batch assays is not recommended
because other nontoxic closures, such as glass beakers, can be uBed
without extensive toxicity studies.
5. Carbon limitation can reduce the X in high-nutrient concentration
samples with batch assays, hence ventilation or aeration (with or
without COz addition) is recommended for batch assays because the
use of specific surface area to volume ratios does not insure that
carbon limitation will not occur.
6>. When aeration or specific surface area to volume ratios are used
in batch assays, the pH must be kept between 7. 0 and 8. 3. With
ventilation or aeration this can be accomplished by increasing or
decreasing air flow rates or by addition of COz to the air as required.
7. The use of suspended solids, total cell volume, or cell numbers
appears to be adequate for parameters of biomass; however, the use
of cell numbers is not recommended as a sole parameter of biomass
because Selenastrum capricornutum cell size decreased linearly with
an increase in medium concentration. Also, at low cell concentrations
or suspended solids levels (< 10 mg/i) the use of suspended solids is
not reliable because of the lack of precision at that level.
8. The satisfactory precision (coefficient of variations 15%) of batch
assay algal responses obtained in the interlaboratory precision tests
11
-------
indicate that this test can be recommended as a screening procedure in
eutrophication analyses.
NATURAL WATER ASSAYS
9. The batch assay method as described in this report is recommended
for the screening of natural waters because linear relations were
observed between & responses and the concentrations of both Clear
Lake and Sacramento River waters, and the & and determinations
had an adequate precision (coefficient of variation 5 18. 5%).
10. The use of yield coefficients in batch cultures is not recommended
for the identification of growth-limiting nutrients.
11. The measurement of in wastewater—receiving water mixture
assays can identify toxicity which is not detected by X responses;
therefore, this measurement is strongly recommended as a general
algal response measurement when batch assays are used.
12. The use of enrichment (spiking) procedures with batch assays to
identify growth-limiting nutrients yielded questionable results in this
investigation; however, the method appears to have merit but further
research and evaluation are needed before these procedures can be
recommended for practical use.
PRESENT STATUS OF THE BATCH ASSAY
13. The batch assay, as described in this report, was inferior in
most respects to the chemostat assay and should be used primarily
as a screening or monitoring method in eutrophication analyses.
14. Further research must be undertaken on methods to identify the
growth-limiting nutrients, to obtain algal growth kinetics of important
nuisance algae, and to develop the rational framework necessary to
apply batch bioassay results to eutrophication problems in a rational
and practical manner.
PRESENT STATUS OF CHEMOSTAT ASSAYS
15. The chemostat assay is the only procedure (method) recommended
for determination of any kinetic description of algal growth—nutrient
concentration relationships.
16. The chemostat bioassay procedure is highly recommended for
determining the concentration of rate-limiting nutrients in natural
waters and wastes, the toxic characteristics of wastes, and for identi-
fying the growth rate-limiting nutrients.
12
-------
17. The use of algal growth kinetic data, and particularly accurately-
determined values of Kg and fx in combination with enrichment
(spiking) tests, is recommended as the most accurate and reliable
procedure for identifying the growth-limiting nutrients in natural
waters or wastewater mixtures,
18. The most rational application of bioassay results to practical
eutrophication problems utilizes kinetic descriptions of the alga and
nutrient of concern in conjunction with the results from chemostat
assays and nutrient concentration assays on the natural water as
described in this report. Its use and further refinement and the
research related thereto are strongly recommended.
19. The most urgently needed information in eutrophication analyses
today is kinetic descriptions ((2, Ks, Yn, k(j, etc. ) for the test alga
as well as troublesome bloom algae and the nutrients presumed to be
of concern, but particularly phosphorus, nitrogen, carbon, and
micronutrients.
20. A biochemical evaluation is recommended strongly to investigate
the phenomenon of "excess" nutrient uptake, particularly that of
phosphorus with Selenastrum capricornutum and to determine accurately
the different forms of phosphorus inside the algal cells and their avail-
ability for growth and storage and the kinetics related thereto.
13
-------
SECTION III
INTRODUCTION
THE PROBLEM OF EUTROPHICATION
Eutrophi.cati.on is the natural or arti.fi.ci.al enrichment of water systems
with nutrients resulting in substantial increases in the productivity and
organic matter (mainly algal cells) of the systems. The increased mass
of organic matter resulting from the increased productivity frequently
limits some of the uses of such water. Activities of man accelerate
the addition of nutrients thereby increasing the productivity. Such
"cultural eutrophication" may be caused by discharge of domestic wastes*
runoff from agricultural operations, wastes from mining and industrial
operations, and disruption of natural drainage basins through develop-
ment, road building, and urbanization. The deleterious effects of
cultural eutrophication have been documented for lakes in this country
as well as abroad [ 1—5], and it has been estimated that over half of the
surface water supplies in the U. S. are already affected by cultural
eutrophication [ 6]. Stewart and Rohlich [7], the National Academy of
Sciences [8], and the Uppsala Symposium [ 9] have reviewed eutro-
phication recently.
Effects of Eutrophication
The main effect of eutrophication is an increase in the organic content
of the water. In eutrophic waters the development of masses of plank-
tonic algae (primarily blue-green algae) has been the focus of much
attention. However, the growth of aquatic plants should not be ignored
nor assumed to be negligible [ 10, 11].
The algal standing crop or concentration in a given water body is a
complicated function of many variables. A simplified conceptual descrip-
tion of the algal standing crop in a natural aquatic system is shown in
Figure 1. Such a description permits identification of those variables
amenable to control by man. Variables which increase the standing
crop are the algal growth rate, the concentration of algae in the input
water, and the rates of resuspension of planktonic and benthic algae.
Variables which decrease the standing crop are the loss of algae in the
outflow, sedimentation of algae, and the algal decay rate which includes
respiration, mineralization, and consumption by predator organisms.
A constant algal standing crop occurs when the rates of the variables
increasing the algal concentration is balanced by the rates of the factors
decreasing the algal concentration. An increase in the standing crop,
which at noxious levels would constitute a bloom, results when the
cumulative rate of the increasing factors exceeds that of the decreasing
15
-------
Change in
Stand ing
Algal Crop
Input of
Algal Cells
Output of
Algal Cells
4
Growth of
Algal Cells
Predation
of Algae by
Higher Organisms
Decay of
Algal Cells
Settling of
Algal Cells
+
Resuspens ion
of Algal Cells
+
Suspension of
Benthic Algae
dX _ F
V o
dt v Xf» ~ V + X5 ) ~ X2
(»)
k - S X, + S X4 + SD Xs
d 1 s 1 r B 5
Where,
V
F
X
o
SB
X„
= Growth rate of algal mass,
= Volume of water body,
= Inflow and outflow rate of water body,
= Algal concentration in inflow water,
= Suspension rate of benthic algae,
= Concentration of benthic algae,
Xj = Planktonic algal concentration in water body and outflow,
X2 = Concentration of higher organisms responsible for algal predation,
X3 = Algal cell concentration available for predation by higher organisms,
X4 = Concentration of planktonic algae on the bottom of water body,
FIGURE 1. ALGAL GROWTH MODEL FOR NATURAL AQUATIC SYSTEMS
= Rate of predation of algal cells,
Sg = Net settling rate of algal cells,
Sf = Resuspension rate of algal cells,
k^ = Decay rate of algal cells,
-------
A decrease in the standing crop, one of the aims of eutrophicatLon
control, results when the cumulative rate of the decreasing factors
exceeds that of the increasing factors.
Control of Eutrophication
Man can control only some of the factors which affect the level of the
standing crop. For example, the rates of predation, sedimentation,
and resuspension are very difficult, if not impossible, to control be-
cause these are functions of variables such as the algal population
composition, higher organism population composition, temperature,
wind velocities, etc. The only way to control the algal standing crop
at present is to reduce the growth rate of troublesome algae.
The growth rate of algae is controlled by environmental conditions such
as temperature and light intensity and also by the concentration of some
rate-limiting nutrient (Liebig's law of the minimum) if all nutrients are
not in excess. Control of the physical environmental conditions affecting
the growth rate is virtually impossible; therefore, control of the algal
growth rate can only be effected through reduction of the rate-limiting
nutrients present in a specific body of water. The development of
methods to remove nutrients from wastes and to determine which wastes
require treatment is dependent upon knowing which nutrients control
productivity and how the concentrations of th.e nutrients affect the growth
of algae. Therefore it is necessary to understand how eutrophication
and the resultant productivity are brought about before rational solutions
to the problem are possible.
The lack of understanding of the mechanism of eutrophication is currently
exposed in the vigorous controversy over phosphorus, carbon, or nitrogen
being the principal limiting nutrient causing algal blooms during cultural
eutrophication [ 12—23], One popular line of thought is that phosphorus
is the principal nutrient involved in causing blooms and it should there-
fore be removed from at least detergents [ 14, 19* 20]; however, there
are other views being expressed that carbon or nitrogen are the nutrients
of concern [ 15, 17, 18, 22—24],
In general, it is accepted that there is a relationship between dissolved
nutrients and the concentrations of algae In a water; also that higher
levels of dissolved nutrients result In greater populations of algae. How-
ever, few investigators have attempted to use or develop the nutrient
concentration — growth rate relationships embodied in fundamental growth
models such as reported by Monod [ 25]. Caperon [ 26], Dugdale [27],
Eppley et al. [ 28],.. and others have used kinetic models to describe
growth and nutrient uptake of marine phytoplankton. Smith [ 29]
suggested a similar approach in modeling ecosystems. McGauhey et al.
[ 30—33] and others [ 34] have used kinetic models in describing studies
of algal growth in freshwater environments and in evaluating batch and
chemostat algal assay studies. In developing rational bases for esti-
mating the allowable concentrations of nutrients in aquatic environments,
17
-------
such studies are necessary for the specific algae of concern, and par-
ticularly for algal species used in standardized algal bioassays.
THE NEED FOR BIOASSAYS
To detect the factors contributing to algal blooms it is necessary to
bioassay both wastewaters and receiving waters. The bioassay relates
the biological response directly to the water and its nutrient content.
Therefore an algal bioassay, as compared to a chemical analysis of the
waters, should provide more knowledge of the bio stimulatory (or toxic)
effects of such waters. Also, a single analysis with a bioassay may
suffice to determine gross biostimulant concentrations, whereas a wide
variety of chemical analyses would be required to predict the enrichment
characteristics based upon the specific nutrient concentrations present.
The need for standard bioassay procedures was recognized by the Joint
Industry—Government Task Force on Eutrophication. Candidate methods
to measure bio stimulatory response were considered at a meeting in
Chicago in March 1968 which resulted in a program to develop and/or
evaluate a test or tests which could be used in the laboratory to assess
quantitatively the algal growth characteristics of a sample of water
[ 35]. Specific criteria for selection of possible procedures were listed
at that time; i. e. , it was agreed that the assays should have the following
characteristics; 1) simple and convenient enough to be performed by
technicians, 2) equipment requirements simple and economic, 3) results
obtained accurate and reproducible, 4) geographical variation in the
method minimal, and 5) results applicable with judgment to the natural
environment.
Three candidate tests were suggested for study with reference to the
foregoing criteria; the bottle test (batch culture), the chemostat test
(continuous flow culture), and an in situ test, the l4C uptake assay
(primary productivity).
BIOASSAY OBJECTIVES
Fundamental differences between laboratory bioassays and the field
bioassay (14C) required that two different approaches be used. First,
a comparative evaluation of the bottle and chemostat tests was to be
made and subsequently a field evaluation of the accepted techniques was
to be conducted. The general objective of the laboratory evaluation was
to compare the two bioassay techniques — batch and continuous flow —
for assaying the algal growth and nutrient relationships in natural and
enriched waters. Part of the general objective was accomplished as
reported in the First Annual Report [ 34J.
The successful application of bioassays to eutrophication control requires
the fulfillment of a number of specific objectives as set forth in the
First Annual Report. In general, it is required that precise measurement
18
-------
of algal growth responses be made followed by the application of these
responses within a rational framework to obtain the desired results.
Therefore, the development and comparison of different bioassay-
techniques need the co-development of a rational framework in which
bioassay results can be applied, with judgment, to solve eutrophication
problems.
The development of a rational framework for application of bioassay
results must be based on the growth rate—nutrient concentration rela-
tionships of algal species of concern. In a specific case the algal
species of concern may differ from those of another case. Therefore,
the growth rate-nutrient concentration relationships (growth kinetics)
of many different algal species must be determined. At the very least,
the growth kinetics of the algal species used as test organisms in
standardized algal bioassays must be determined to enable interpreta-
tion of algal responses in such bioassays.
PREVIOUS ASSAY TECHNIQUES
Many scientists have developed specialized bioassay techniques to deal
with specific water and productivity problems. However, there has
been little agreement in the data obtained with different techniques and
it is difficult, if not impossible, to relate the results obtained from the
various methods. Some examples of the different bioassays are dis-
cussed to illustrate the diversity of methods used and to indicate possible
directions of research on bioassays.
Several investigators have utilized physiological measures as indicators
of limiting nutrient concentration and productivity in field samples.
Fitzgerald [36], in studies on natural populations of algae and vascular
plants, showed that comparative uptake rates for ammonia and measure-
ments of surplus phosphorus or alkaline phosphatase activity can be
used to indicate the relative nitrogen or phosphorus availability to the
cells. Fitzgerald's [ 36] results showed that non-nitrogen-fixing algae
will take up ammonia at a rate indicating that nitrogen is not readily
available. However, in the same environment phosphorus analyses
indicated that available phosphorus was in short supply for nitrogen-
fixing algae. Therefore, different nutrients can be limiting in the same
environment for different species of algae having different requirements.
Fitzgerald [ 36] pointed out the sampling difficulties involved in these
bioassays and also indicated some of the important environmental factors,
such as rainfall and circulation currents, affected the observed relation-
ships. Eppley and Coatsworth [37] studied uptake rates of nitrate and
nitrite and suggested that it may be possible to utilize these results for
developing bioassays similar to the ammonia analysis of Fitzgerald [ 36].
Holm-Hansen in a series of papers, e.g. [ 38], indicated that ATP,
DNA, and organic carbon measurements are useful indices of algal mass.
DNA and organic carbon measurements were closely related but did not
19
-------
distinguish between dead or near dead plant material and live cells
as did the ATP test-
Chlorophyll "a" measurements have been used to indicate the activity
of algal populations (e. g. , Strickland and Parsons [ 39]). The develop-
ment of an "in vivo" chlorophyll "a" technique for the analysis of mixed
populations of phytoplankton in the field proved to be convenient and
sensitive [40], Because chlorophyll varies with light and physiological
state of the algae, measurement of chlorophyll might indicate the state
of the cells, that is physiologically active or Inactive, as well as pro-
viding rough estimates of algal biomass. As an application of chlorophyll
measurements, Manny [41] used pigment ratios (chlorophyll "a" and
total carotenoids) of freshwater and marine phytoplankton to indicate
nitrogen deficiency.
The most commonly used techniques for field bioassays are productivity
measurements, The light and dark bottle methods of determining
photosynthetic rates (either oxygen production or carbon uptake) have
been used for many years, primarily in marine waters but also more
commonly in recent years in the freshwater environments. The first
productivity measurements were the oxygen bottle techniques of Gaarder
and Gran [42]. Later a more sensitive technique was devised by
Steemann-Nielsen [ 43] using the radioisotope I4C and measuring the
uptake of UC02. This latter method achieved considerable prominence
as a means of measuring productivity m situ. Goldman [ 44] advocated
the use of the 14 C technique in freshwater environments. Although
investigators do not know exactly what phase of algal growth the 14CC>2
uptake measures (generally it is considered "net photosynthesis"), it
appears to be a convenient and sensitive tool for measuring productivity
in natural phytoplankton communities [39] and in determining trace
element requirements in freshwater lakes. The in situ analysis suffers
from the disadvantage that variations in the algal population sampled
may produce results which are difficult to interpret in terms of the
biostimulatory response of that water.
Considerable effort and attention has been devoted to laboratory bioassay
analysis of water samples. Historically the major method used has
been a batch culture assay. However, recently attention has been
directed to the use of continuous flow culture bioassays. Both techniques
use an essentially constant volume of culture differing operationally
only in that no flow occurs in batch assay and a continuous and constant
flow of fresh sample occurs in the continuous system. In batch systems
the cell and nutrient concentrations change with time while in continuous
systems a steady state is attained (see Section IV for further discussion).
Because of the traditional use of batch culture assays and their adapt-
ability to the handling of large numbers of samples, batch culture
methods have been used extensively to date.
Allen and Nelson [45] first used the batch technique to evaluate the
capacity of -water samples to grow algae. In recent years investigators
20
-------
have used either the response of unialgal cultures [ 46—48] or of mixed
algal cultures [ 49, 50] to determine the level of enrichment in water
samples. Although unialgal cultures are easier to use, mixed culture
advocates believe that a mixed species community approximates the
natural situation more closely. Although generally the algal biomass
is estimated (weight, numbers, or volume), estimates of algal growth
of mixed algal species have been made using chlorophyll fluorescence
measurements [50] or 14C accumulation [ 5 1 ]..
Frequently the results from batch bloassays^of water samples are
expressed as maximum cell concentration, X, sometimes called the
maximum yield of cells. Although this relationship may be reproducible
for a given sample and fixed set of conditions and may give an estimate
of the total amount of cells which can grow on the nutrients present in
the sample, that value is far removed from the level of algae (standing
crop) which occurs in nature. Obviously the fact that one is able to
demonstrate growth in excess of that of the natural community in water
samples indicates that the maximum cell concentration, X, is rarely if
ever attained in nature. Moreover, one would expect that the effects
of environmental limiting factors (light, temperature, mixing) and the
effects of predation and settling of cells would tend to keep the con-
centration of algae below the maximum possible level. Since the effects
of environmental factors are most frequently expressed in terms of
response rates, it is logical that the bio stimulatory response be ex-
pressed in terms of growth rates also. Response parameters such as
maximum growth rate and maximum cell concentrations are being
evaluated in ongoing-studies at Lake Tahoe to assess biostimulatory
response and to determine which parameters are best In terms of
accuracy and precision as well as practical Interpretabillty [ 30—34].
However, no definitive results have been obtained to date. One of the
major reasons for the inconclusive results appears to be the inherent
variability of batch assays.
Use of the chemostat, a continuous flow culture device, as a bloassay
tool has been relatively limited. In the chemostat, cultures are main-
tained at steady state where all factors are at a steady state, i. e. ,
nearly a constant level. Thus the determination of growth constants is
facilitated through the use of cultures in balanced growth. The merit
of the chemostat for nutrient-growth relationships was reviewed by
Pearson [ 52] and the results of bloassays of natural waters in the
Lake Tahoe Basin and of sewage effluents diluted in these natural waters
[ 30—33, 53] have Indicated that It should be evaluated thoroughly for
use as a routine laboratory assay. Pearson et al. [ 53] have described
some comparative bioassays of various waters using batch and chem-
ostat techniques and have suggested that the batch technique produces
deceptive results. Borchardt in ongoing work is using the turbidostat
(the same fundamental theory as the chemostat) to study natural water
samples [ 54].
21
-------
Application of results obtained from laboratory bioassays of samples
collected from natural waters or waste effluents will always be limited
by the degree to which the bioassay models reflect what happens in
nature. As soon, as a sample is removed from its normal environment
one has introduced factors which make that sample nonrepresentative.
The effects of light (period and cyc]e), temperature, the hydromechanics
of the aquatic system, exchange of dissolved gases with the atmosphere
or benthos, and the presence of decomposers and predators all have
been altered because these factors are properties of the environment
and are changed simply by the physical fact that the sample is removed
from that environment. Therefore, one can only analyze for those
properties which are intrinsically a part of the sample, its chemistry
(concentration of nutrients and toxicants), and the concentration and
composition or organisms, detritus and other particulate matter.
Further treatment of the sample such as by filtration, to remove partic-
ulate matter so that a standard unialgal population may be introduced
may only serve to make more difficult the extrapolation of laboratory
results for application to nature.
The analyses of field samples suffers from the disadvantage that they
measure only what is in the particular sample at the collection time.
Also, field bioassays do not allow ready interpretation of the effects of
various treatment or of additions of wastewater. Therefore one virtually
is forced to bring samples to the laboratory to obtain such information.
Although the basic techniques of laboratory bioassays are similar, very
little effort prior to the Provisional Algal Assay Procedure evaluation
study has been devoted to standardize methods. Hence, the degree of
standardization of algal assays depends in major part on the progress
made in this evaluation,
PROVISIONAL ALGAL ASSAY PROCEDURES EVALUATION
To develop and/or evaluate and standardize an algal assay technique
many different problems must be investigated. In the evaluation of the
Provisional Algal Assay Procedures by the participating groups much
attention has been given to the methods for measurement of algal bio-
mass [34, 55—58] and the batch culture assay [34, 55—72]. However,
the evaluation of the continuous culture assay method has received
relatively little attention [73].
The algal biomass methods that have been studied include measurements
of algal cell numbers (hemocytometer counts and electronic particle
counts), dry weight, total cell volume, optical density, fluorescence,
total carbon, and suspended carbon. In general, good relationships
were found between different biomass parameters [34]; however, some
measurements appear questionable. Cell numbers appear to be un-
satisfactory [34, 55, 61] and fluorescence measurements may be suspect
under nitrogen-limited conditions [74], The use of more than one
method during assays is recommended, with all measurements
referenced to cell mass on a dry weight basis [73],
22
-------
The batch assay method has been shown to respond to varying levels of
nitrogen and phosphorus [ 34, 59> 60, 70, 72]. The need for the macro-
nutrients, calcium, potassium, and sulfur [ 60,70] and carbon [ 75] was
demonstrated, and a sensitivity to iron [70] and other micronutrients
(included as a mixture) was also shown [ 57, 61, 62, 67, 76]. Therefore
the inclusion of micronutr ients in the reference media was recommended
[ e. g. , 57, 61, 76].
The coefficient of variation of the batch culture assay method appears
reasonable at approximately 15% [ 34]; and similar coefficients of
variation (10% — 30%) were observed in an interlaboratory precision
test of the batch assay [69]. Carbon-limitation effects were noted in
assays of high-nutrient concentration samples [ 55, 5 7, 65] resulting in
nonlinear responses between concentrations of nutrients (other than
carbon) and algal growth [ e. g. , 55], Therefore, it was recommended
that the pH of the cultures during assays be kept below 8. 5 through use
of appropriate surface to volume ratio flasks [55] or ventilation systems
[ 34, 55], Light intensity can influence the growth of the test algae [ 55];
however, the use of low intensity lighting (150 ft-c) has been suggested
[71], No significant differences were observed between higher light
intensities (400 and 800 ft-c) when cell number was the parameter of
biomass; however significant differences were observed when carbon
content was the parameter of biomass [ 58].
Nutrient utilization rates in batch assays were found to be influenced
by the presence ox absence of micronutr ients [ 58], and excess nutrient
uptake and initial N/P ratios influenced the organism yield coefficients
for nitrogen and phosphorus [ 34]. Excess uptake of nutrients appears
to have occurred in other studies [ 59. 68]. Inoculum age (between 4
and 35 days) had no apparent effect on organism, response in batch assays
[ 58, 61, 71]. Identification of the growth-limiting nutrient in samples
in batch assays through the use of enrichment (spiking) techniques
succeeded in some cases but not in others [ 72], Further study is
needed.
The batch assay method has been evaluated extensively. Although the
batch assay cannot fulfill all requirements of algal assays, it is expected
that a standardized assay method will be recommended in the near
future [77]. However, the chemostat assay has not been evaluated and
this still remains to be accomplished [73], Porcella et al. [ 34] and
Toerien et al, [ 78] have discussed the application of chemostat assays
in eutrophication analysis. The potential dual role of chemostat methods,
both as an algal assay method and as a method of obtaining the growth
kinetics (growth rate—nutrient concentration relationships) of algae was
pointed out [78],
Determination of the growth kinetics of both the algal species used in
standardized algal assays and for those algae important in eutrophication
is needed to allow interpretation of test organism behavior as well as
rational design proposals to solve eutrophication problems. For
example, the low half saturation constant (K) for phosphorus observed
s
23
-------
for Selenastrum caprtcornutum [ 34] explains why the growth rates of
this organism do not differ significantly from each other under condi-
tions for growth having wide variations in the phosphorus concentration.
That is, the phosphorus concentration is so high (excess P) that it is
not rate limiting in most instances; hence the growth rate is zero order
(independent) of phosphorus concentration in this range. Furthermore,
the low values of K for phosphorus suggest that for this organism (and
types behaving in a similar manner) phosphorus concentrations will
have to be extremely low (< 20 \xg/£) to be the growth rate limiting
nutrient in nature. It may not be feasible or possible in many waters
to reduce the danger of blooms from these types of organisms by re-
ducing background phosphate concentrations to such levels by domestic
waste treatment for phosphate removal.
REPORT OBJECTIVES
The specific objectives of this research investigation and report can
be summarized briefly as follows:
1. To evaluate the use of batch culture assays with a standardized,
chemically-defined medium.
2. To evaluate the use of batch culture assay of natural samples.
3. To evaluate alternative methods for sample pretreatment.
4. To determine the growth kinetic constants for the test algal
species, Selenastrum capricornutum, in a phosphorus-limited environ-
ment.
5. To evaluate the use of chemostat assays of natural samples.
6. To compare the relative efficacy of batch and chemostat assays in
eutrophication analyses.
24
-------
SECTION IV
THEORY AND RATIONALE
GENERAL THEORY OF GROWTH
OF MICROORGANISMS
Symbols used in the development of kinetic equations in the following
pages are based upon the unified concept of fundamental symbols sug-
gested by the "Second International Symposium on Continuous Cultiva-
tion of Microorganisms" held in Prague, 1962 [79]. The symbols,
their units, and explanations are summarized in Section XL
Growth of Unicellular Organisms
In the biological sense, growth is an orderly increase of all the
components of an organism, not merely of some of its constituents.
In all cellular organisms, cell multiplication is a consequence of
growth. In unicellular bacterial and algal growth, cell multiplication
leads to an increase in the number as well as mass of cells.
Generation Time and Growth Rate
Because growth represents an increase in the amount of living matter,
it leads to an increase in self-duplicating material. The more living
substance that is formed, the greater becomes the amount of matter
that can grow and reproduce itself. Consequently, under ideal con-
ditions for growth and reproduction, the amount of living matter
increases not in direct proportion to time, but it multiplies itself by
a constant factor in each successive unit of time. The geometric
progression with time is illustrated most dramatically by the growth
of cultures of most unicellular bacteria and algae. Each individual
cell grows, and after reaching a certain size, divides to form two or
more complete individuals. In cultures the time of division is ran-
domized, but the time required for each successive doubling of the
total population under optimum conditions remains constant. This
period of time is called the generation time (g).
Because the number of cells doubles during each generation, the total
population increases as the exponent of 2, and such growth is referred
to as exponential growth and can be algebraically described as:
X = X 2N (i)
o
25
-------
where
X = concentration of microorganisms at time t, M/V,
XQ = concentration of microorganisms at time 0, M/V, and
N = number of cell divisions
The following equation
t
N " 8 <2>
where
t = time, t,
N = number of cell divisions in time interval t
g = generation time, t
therefore expresses the average generation time or doubling time for
the culture as a whole [80], A rearranged form of Equation 2 can be
substituted in Equation 1 and one obtains
X = (X0)(2t/g) (3)
Taking logarithms and rearranging
a = (ln 2> (t) (4)
g In (X) - In (XQ)
and solving for X,
In (X) = t ^g-2i + In (Xo) (5)
When the expression (In 2)/g reaches its maximum value for a specific
organism under specific growth conditions, the maximum specific
growth rate, jx, can be substituted for (In 2)/g as follows:
In (X) = jxt + In (XQ) (6)
or
X = (XQ) e^ (7)
26
-------
The equation in this form is recognized as the general growth equation
[80], Taking the derivative with respect to time, the following
relationship is obtained
ft = ^ <8>
The gross specific growth rate is less than the maximum growth rate
when either the substrate is limiting or when conditions are not
optimum. The net growth rate is obtained when both the specific
growth rate (|x) and organism decay rate (k^) (i. e, , the decrease in
cellular mass due to endogenous respiration and cellular death with
subsequent lysis) are considered:
dX
dt
(M- -
kd)X
(9)
Specific Growth Hate as a Function of
Limiting Nutrient Concentration ~~"
The specific growth rate, (J-, is specific for organism, environment,
and culture medium. It depends on the growth capacity of the micro-
organism and the environment in which it grows. The specific growth
rate is a function of the growth rate limiting nutrient concentration
when all other factors are in excess. Then
H = f(S) (10)
where
S = growth rate limiting nutrient concentration, M/V.
This function is generally expressed in such a way that atlow nutrient
concentration the specific growth rate is first order with respect to
nutrient concentration, whereas at high nutrient concentration, the
specific growth rate is zero order with respect to nutrient concentra-
tion and it approaches its maximum value,
Monod Model (Michaelis-Menten Equation)
The rectangular hyperbola is the model widely accepted in expressing
the relationship between specific growth rate, ji, and the rate-limiting
concentration. This relationship is based on the Michaelis-Menten
equation which was developed to describe the rate of an enzyme reaction
as a function of substrate concentration [81] and which was proposed
27
-------
empirically by Monod [25, 82] to describe the relationship between
bacterial growth and substrate concentration:
where
(jl = maximum specific growth rate, time 1
fx = specific growth rate, time"1
S = concentration of the growth rate limiting nutrient, M/Y
K = half saturation constant, numerically equal to the
nutrient concentration at which the specific growth
rate is one-half the maximum growth rate, i. e. ,
H = p/ 2.
The Michaelis-Menten (Monod) kinetic model for describing the relation-
ship between the growth rate and nutrient concentration is shown in
Figure 2, and this relationship has been widely used by microbiologists
and engineers working with continuous culture systems [52, 80, 83—85].
In sanitary engineering, several workers have been successful in
using this model for descriptions of activated sludge kinetics, anaerobic
fermentation system kinetics, and also algal growth kinetics. Caperon
[26], Dugdale[27], Eppley et al. [28], and others have used this
kinetic model to describe growth and nutrient uptake of marine phyto-
plankton, McGauhey et al. [30—33] have used the same kinetic model
in descriptive studies of algal growth in fresh water environments.
Porcella el^al^ [34] used the M-M, M equation in evaluating batch and
chemostat algal assay studies.
Other Empirical Kinetic Models
Several empirical kinetic relationships have been proposed to describe
growth kinetics of microorganisms. However, none of these models
had the wide acceptance of the M-M, M-kinH.it- model,
Teissier's Exponential Model
Teissier [86] proposed the following exponential equation to describe
the growth rate-nutrient concentration relationship of his bacterial
system
^ ' -kS \
M- = H- i.1 " e / (12)
where
k = a growth constant.
28
-------
GROWTH LIMITING NUTRIENT CONCENTRATION, S,
FIGURE 2. A GRAPHICAL COMPARISON OF MICHAELIS-MENTEN,
(MONOD) EQUATION, EXPONENTIAL EQUATION AND
BLACKMAN MODEL WITH THE SAME VALUES OF
MAXIMUM SPECIFIC GROWTH RATE AND HALF
SATURATION CONSTANT
Schulze and Lipe [87] and Schulze [88, 89] found that their results fit
the exponential equation better than did the M-M, M model. Shelef
et_al. [90] in studies of algal growth kinetics under nitrogen-limiting
conditions, suggested a modified Teissier's equation:
. / "S* /Ka\
M- = M- (1 - e 1 (13)
29
-------
in which Ka is a growth constant equal to the value of nutrient concen-
tration at a specific growth rate of 0, 632 jl. A graphic comparison of
this method is also shown in Figure 2,
Blackman First Order—Zero Order Model
The Blackman first order—zero order kinetic model is based on the
assumption that the rate of photosynthesis is proportional to the light
intensity or carbon dioxide concentration until a maximum rate is
attained at a certain limiting concentration (or light intensity) and
thereafter the rate remains constant at the same maximum level
regardless of any increase in nutrient concentration or light intensity.
This model can be formulated as:
fx = Kc S, when |x < jx, S < Sc
|x = (x, S = to, when |x = jx, S £ Sc
where
S = nutrient saturation concentration, M/V.
c
A graphic comparison of this model is also shown in Figure 2.
Moser's Model. Moser [91] developed a nonrectangular hyperbola to
describe the relationship between growth rate and the concentration of
the rate-limiting nutrient as shown in the following:
H- = £( nl (14)
Ks + S
In case of n = 1, Moser's equation is the same as the M-M, M
equation.
Contois' Function; Contois [92] proposed a kinetic equation for
description of the specific growth rate in terms of population density,
P, as well as the concentration of the limiting nutrient, S:
* 3 ^(bp + S ) (15>
where
B = growth parameter which is constant under defined
conditions.
30
-------
Growth. Yield
The cell yield coefficient, Y, is defined as the amount of cell mass
produced per unit amount of nutrient depleted in the culture system,
and is expressed as,
Y _ dX Cell mass produced
-dS ~ Amount of growth-limiting nutrient removed
It is generally believed that the yield coefficient is a constant. How-
ever, the yield coefficients for nitrogen-limited and phosphorus-
limited growth of algae have been shown to be quite variable [34, 93—
102], Generally, yield coefficients are expressed in terms of the
rate-limiting nutrient; however, it should be noted that they can be
expressed in terms of any constituent of the substrate consumed by
the organisms.
GROWTH KINETICS OF MICROBIOLOGICAL SYSTEMS
Ideal Model Reactors
There are three types of systems, classified in terms of flow condi-
tions, which are commonly considered as ideal reactors. These are:
Batch or Nonflow Mixed Reactors. In the typical batch reactor the
reactants, viz. nutrients and algae, are charged initially into the
reaction vessel, mixed thoroughly, and retained for a period of time
during which their interaction proceeds. The end results of the
reaction are increased quantities of cells and decreased quantities of
nutrients in the nutrient solution. The batch system is a nonsteady
state system in which both nutrient and algae concentrations change
continuously with time; however, at any given instant these concen-
trations are uniform throughout the reacting mass.
Plug Flow or Nonback Mix Flow Reactor. In a plug flow reactor each
element of the flowing mass follows another in sequence as though
separated from it. Consequently, there is no longitudinal mixing
between the elements although there may be lateral or radial mixing
within each element. Thus a plug flow reactor may be considered as
consisting of an infinite train of batch reactors, each having the same
residence time as the plug flow reactor itself. Consequently, the
kinetics of batch and plug flow reactors are the same.
31
-------
Continuous Flow Stirred Tank Reactor (CFSTR) or Back. Mix Flow
Reactor. A CFSTR is a simple homogenous open system consisting
of a stirred tank into which growth medium is fed and from which
partially expended culture medium emerges at the same rate. The
volume is maintained constant by some form of constant level take-
off. In the CFSTR the feed rate is maintained at a predetermined
level and the effluent containing reaction products (including algal
cells) is withdrawn at this same rate. Thus the physiological condi-
tions of cells in the reactor are maintained at a predetermined level
and the entire system (nutrients and cells) is at a steady state.
Algal Growth Kinetics in Batch Systems
Batch Growth Curve (Unbalanced Growth). The basic concern with
algae in eutrophication is their multiplication or growth. The dynamics
of multiplication, as well as of other functions related to it, are the
most important aspects of algal growth. Figure 3 represents an ideal
batch growth curve in which cell concentration-time, and nutrient
concentration-time profiles are superimposed. The algae begin
multiplication in a surplus of nutrients, which gradually decreases;
however, the single components often are utilized at different rates.
After inoculation of the nutrient medium, some time elapses before
the number of algae begins to increase. This period is termed the
"lag phase" (Zone i, Figure 3) and the length of time before cell
division depends on the type and age of inoculum (phase of growth of
the culture from which the inoculum originated) and the characteristics
of the medium. This dependency on tEce age of the inoculum reflects the
wide physiological variation between different growth phases in an
ordinary algal batch culture toward the end of the lag phase. The mass
of algal cells present in the culture increases considerably without cell
division occurring. The cells are preparing for cell division and multi-
plication. This stage is therefore called the "physiological youth" of
the culture, a description which must be taken figuratively, as it is
not possible to consider the culture as a single organism. The next
stage represents the beginning of multiplication and of the exponential
or logarithmic part of the growth curve (Zone 2, Figure 3) in which
the algae are multiplying uniformly with an approximately constant
growth rate. The rate of growth during this phase is characteristic
for any given alga under the particular condition of cultivation and
represents the maximal reproductive capacity of that alga in the
specific environment. The environmental factors that govern the rate
of growth include the nature and concentration of the nutrients, pH,
temperature, light, and other physical and chemical variables. The
length of the exponential phase differs according to the available
amounts of some essential nutrient. However, in a batch culture the
period in which the living biophase does not change its properties and
grow in a dynamic steady state is generally short and often steady state
does not occur at all but is replaced by continuous transition states.
32
-------
0 LAG PHASE
-------
culture method. Therefore, the exclusive use of batch cultures over
other methods should not be advocated because of their "unbalanced"
growth characteristics.
Evaluation of Batch Growth Kinetics. Mathematical expressions for
the kinetics of growth in batch cultures have been presented in the
section on the general theory of growth of microorganisms. The
growth parameter of batch assays of water samples traditionally is
expressed most often as the maximum cell concentration, 5£. It is
presumed that 5c allows differentiation between samples based on
initial nutrient concentration. However, due to difficulties in inter-
preting what actually limits X and in the application of & values to
practical problems, attention has been directed to the use of other
growth parameters, such as maximum specific growth rate batch,
jjL-j.,; half saturation constants, Ks, for each rate limiting nutrient;
and yield coefficient, Y, for each nutrient,
A A
Maximum Cell Concentration, X. The X from batch cultures can be
described using the following equation,
X = Xq (7)
where
t = length of time needed to reach zero growth rate
and is measured experimentally as the final cell concentration at the
stationary stage of batch growth.
As shown in Equation 7, maximum cell yield varies with initial
inoculum, maximum specific growth rate batch, and time to reach
zero growth rate. The important concept to remember is that the
maximum specific growth rate batch, fL, is a single function of the
growth rate limiting nutrient concentration (Equation 10), while the
time needed to reach zero growth is a complex function of all nutrient
concentrations and possibly other factors.
Maximum Cell Concentration Limiting Conditions. Batch algal growth
in rich "unrestricted" nutrient environments (i, e. , all nutrients are
considerably in excess) is characterized by the situation where the
(iv approaches (I (largely determined by the environmental conditions
of the test) and hence, none of the nutrient species is the growth rate
limiting nutrient. This condition is shown schematically in Figure 4,
Due to the difficulties of identifying the presumed growth rate limiting
nutrient and the possibilities of multiple growth rate limiting conditions,
there is no rational mechanism to describe the relationship between
34
-------
maximum cell production and the concentration of each nutrient species.
However, theoretically a maximum cell concentration limiting nutrient
is defined as the nutrient species which is exhausted first during the
batch growth and presumably has the most significant effect on maxi-
mum cell concentration.
TIME.t
FIGURE 4, BATCH ALGAL GROWTH IN UNRESTRICTED
NUTRIENT ENVIRONMENTS
Maximum Cell Concentration and Growth Rate Limiting Conditions. If
the algae grow under a~specific growth rate limiting condition (i. e. ,
only one specific nutrient limits the growth rate while the other
nutrients, growth substances, and growth factors are in excess), the
only meaningful growth parameter related to the limiting nutrient con-
centration is the specific growth rate as indicated in Figure 5. Although
the rate-limiting nutrient limits the maximum cell concentration as
well, the maximum cell concentration is also related to the length of
time needed to reach zero growth rate. This time period cannot be
35
-------
defined because of its variable character depending upon the character-
istics of the sample, the condition of the test alga, and the test method.
Z
o
5
tr
\—
Z
Ul
o
z
o
o
UJ
o
o
o
TIME.t
FIGURE 5. BATCH ALGAL GROWTH UNDER SPECIFIC
GROWTH RATE LIMITING CONDITIONS
Maximum Specific Growth Rate, To determine the value of the
fj^, equation b can be rearranged, as follows:
% =
In PC , ) - In PC )
1 n+l' * n'
(17)
where
n+x
Xn
t
= cell mass at time n+i
= cell mass at time n
= time interval between exponential growth at
time n and n+i .
36
-------
As stated previously, the is affected by the environmental condi-
tions of the test and is limited by the concentration of the rate-limiting
nutrient (Equation 11)
The specific growth rate batch is equal to the maximum growth rate
multiplied by the fraction S/ (Kg + S). As Ks is always greater than
zero, the value of the fraction is smaller than one and approaches
ji asymptotically as S is increased (i. e. , S » Ks),
To predict the effects of addition of nutrients to and/or removal of
nutrients from a water, one must be able to relate the biostimulatory
response to the concentration of the specific nutrient. If it is deter-
mined that sufficient nutrients are present in the sample to allow (1^
to approach p, it can be concluded that the nutrients are in excess in
the natural situation and some other factor (s) (light, COz, temperature,
predation, etc, ) is limiting the algal population (see Figure 1). On
the other hand, if (I « p,-, some nutrient is likely limiting growth in
the batch culture ana may be controlling growth in nature. Consequently,
{xy should be the most interpretable biostimulatory growth response
parameter from batch cultivation for specific growth rate -limiting
conditions.
Algal Growth Kinetics in Continuous
Culture (Chemostat) Systems
Continuous flow (chemostat) culture is characterized by the continuous
addition of fresh medium (or sample) to the culture, complete mixing
of culture and medium, and a continuous outflow of part of the culture
at a rate identical to the inflow rate of fresh medium, ensuring a
constant culture volume. As long as the chemostat operates, an open
dynamic system results in which a steady state is attained. The algal
cells in the growth vessel continue to grow at the expense of the fresh
nutrients and the total number of algal cells in the chemostat remains
constant. Furthermore, the growth rate of the cells can be controlled
at will because the slower the fresh medium is delivered to the growth
vessel, the slower the cell growth rate. Chemostats provide a method
for keeping an algal culture growing for an indefinite period of time,
at a constant population size, a controlled and constant rate of growth,
a specific, constant physiological state, and constant environmental
conditions. It is obvious that the "physiological state" reflects the
sum of the biochemical, morphological, and especially physiological
features and activities characteristic of the most important factors
operative for growth and metabolic activities of the organisms. It was
noted in the description of the growth curve for batch cultures that the
physiological state often changes rapidly under different environmental
(11)
37
-------
conditions and is the most important characteristic of the state of the
culture. However, with the continuous flow (chemostat) culture,
culture conditions and the physiological state are maintained essen-
tially constant and at a steady state.
Continuous homogenous cultures can be characterized as complex
open systems of constant volume, being as a whole in a dynamic
steady state with essentially constant concentrations of all nutrients
and of cell biomass. During continuous flow cultivation the biophase
permits determination of the kinetics of growth and of growth parame-
ters which can be used in both theoretical analyses as well as for
application to practical problems, as will be shown later.
Basic Assumptions. To develop rationally a mathematical model for
a continuous flow (chemostat) system, three basic assumptions are
made: 1) A constant proportion of the organisms are viable; 2) The
(gross) specific growth rate of organisms in a chemostat is some
function of the growth rate - limiting nutrient concentration, thus
^=Xll = f(S) ; {10)
and 3) The specific growth rate of the organisms varies with the
rate of consumption of the growth rate limiting nutrient,
§ = <18>
where
Y = cell yield coefficient.
It is generally assumed that the yield coefficient, Y, is a growth
constant with respect to any nutrient. However, this is only true
when the energy requirement for endogenous respiration for the
organism does not affect the constant endogenous metabolism of the
organisms at different growth rates, and/or when there is no "excess'
uptake or storage of the growth rate- limiting nutrient [34, 93—102J.
Growth Kinetics in Chemostat with a Constant Yield Coefficient. In
developing the growth kinetics of a continuous flow culture system with
no cell return and with a constant yield coefficient, the continuity of
biomass and nutrients in steady state or the materials balances for
biomass and nutrients are developed.
38
-------
Materials Balance for Cell Biomass. For a CFSTR (Figure 6) a
materials balance for the reactor is expressed by the following
equation:
rate of change of
cell biomass in
the reactor
rate of input
of cells
rate of output
of cells
where
F = flow rate, volume/time
V = volume of reactor.
At steady state (that is dX/dt = 0) in a reactor with no cell recycling
and receiving no cells in the input, Equation 19 becomes
f - *d = v' i = t- ,20>
c
where
0 = hydraulic residence time, t
@c = mean cell age in the reactor, t.
39
-------
/
Symbols <
F = Flow Rate
X = Cell Mass Concentration
S = Limiting Nutrient Concentration
V = Reactor Volume
Subscripts
0 = Inf luent
1 = Effluent or Reactor
FIGURE 6. CONTINUOUS FLOW, STIRRED TANK
REACTOR SYSTEM (CFSTR)
Materials Balance for Nutrient. The materials balance of the rate-
limiting nutrient through a noncell recycling reactor can be described
as follows:
rate of change of
growth rate
limiting nutrient
in reactor
rate of input of
rate-limiting
nutrient
rate of output of
rate-limiting
nutrient
rate of uptake of
rate-limiting
nutrient
dSt
dt
= F S - F Sj - uptake
(21)
The term "uptake" can be defined in terms of the yield coefficient, Y,
and the derived form of the natural growth equation (Equation 18).
Thus,
40
-------
dS ds dX.l Y
dt " dX ' dt ~ Y ^ { ]
and "uptake" = |jlXi V/Y.
Therefore,
V§ = FS - F S. - ^ V (23)
dt o 1 "
At steady state
or
1=0
F (S0 - S, ) = ^ • V
„ F
-------
± = A. = Yq - kd = » - kd
(27)
c
It should be noted that 1/9 = 1 /©c is only true for nonrecycle reactors.
Substituting Equation il into Equation 27 and solving for Sp
Equations 24 and 28 describe steady state values of cell concentration
and limiting nutrient concentration, respecitvely. Inspection of
Equation 28 reveals that for nonrecycle reactors where 1/0 = 1/0C,
once the residence time, 0 is selected, the reactor or effluent concen-
tration Sj has a single value, that is, it is fixed by the residence time
as all other terms in the equation are constants. By substituting the
value of Si from Equation 28 into Equation 24 one obtains the following:
It should be emphasized that at steady state and for a given residence
time 0, and organism, the cell concentration Xj is a function only of
the concentration of nutrient (SQ) in the feed stream. The relation-
ships defined by Equations 24 and 29 are shown graphically in Figure
7, Again it should be noted that for a given residence time, the
effluent concentration Sj, is fixed and independent of the feed concen-
tration SQ. However, the reactor cell concentration (Xj, X2, and X3)
is determined by feed concentration SQ , SQ > and So for that residence
time. There is a critical value of residence time below which the algal
concentration, Xj, decreases exponentially and finally approaches zero
because the residence time is insufficient for the algae to reproduce
and maintain the population. Then the nutrient concentration in the
chemostat and in the effluent, Sj, increases until it reaches the influent
nutrient concentration, SQ, because there is no uptake and no algae
present. This critical value is referred to as "washout. " Using
Equations 20 and 27, when 0 approaches "washout" (0W), jj. approaches
its maximum value, pi, then
(28)
(29)
1 A
r-= - ka
(30)
where
0w = residence time at which "washout" occurs.
42
-------
>
N
s
en"
*
2
O
S5
a:
UJ
o
2
O
o
z
UJ
3
2
UJ
3
_l
ti-
ll.
UJ
o
Q
<1
UJ
cn
RESIDENCE TIME
FIGURE 7. THEORETICAL STEADY STATE CONDITIONS
IN CHEMOSTAT
Growth Kinetics in a Chemostat with Varying Yield Coefficients. Vary-
ing yield coefficients have been observed by many researchers [34, 93—
102], and several explanations have been proposed for this phenomenon.
Marr et_al;_ [99] attributed the varying yield coefficient to the effect of
a maintenance requirement for exogenous nutrient by the organisms.
Hetling et^al^ [ 100] attributed the variation of yield coefficient with
the dilution rate in chemostat bacteria to the maintenance requirement
and presence of dead cells. Mor and Fiechter [ 101, 102] explained
the decrease in cell yield coefficient at low dilution rates as being due
to the effect of a constant endogenous metabolism and increased
inhibitor formation and concentration. A constant endogenous metabo-
lism becomes a large fraction of the total metabolism when the total
metabolism energy requirement is relatively low. In addition, a
varying yield coefficient caused by "luxury or excess" nutrient uptake
or nutrient storage has been reported by several investigations [96—98].
43
-------
However, very little is known about the "excess" uptake mechanism
or the "excess" nutrient uptake function and its effect on growth
kinetics in chemostats.
Nutrient Uptake and Utilization, Two distinct phases can be postulated
for the utilization of an environmental nutrient in photosynthetic cell
growth; namely, the uptake (removal) of the nutrient form the sur-
rounding medium into the cell and then the subsequent role of the
nutrient in the reactions leading to cellular production. This can be
summarized as follows:
g Step 1 /^s Step 2 g
external ^ C internal cell growth,
Cell
where
S = external nutrient concentration
external
S. 1 = internal nutrient concentration available
in erna ce^ growth utilization
S ,, ... = nutrient concentration being utilized for
cell growth ,,
° cell growth.
ii
Excess" or luxury uptake of nutrients is a phenomenon where the
rate of nutrient uptake (hereafter called the uptake rate) exceeds the
rate at which the nutrient is utilized in growth (hereafter called
utilization rate). Consequently the nutrient is accumulated (and
presumably stored) in excess to the amount needed for growth.
"Excess" uptake of phosphorus has been observed for various organ-
isms [34, 96-98, 103, 104].
In batch cultures the uptake rate can be estimated from the nutrient
depletion with time of growth as shown by Porcella et_alj_ [34] for
Selenastrum capricornutum under conditions of different initial N/P
ratios. They found that phosphorus was essentially completely
removed by the algae before 50% of the maximum amount of algal
growth had taken place, indicating "excess" uptake of this nutrient.
They also reported that the wide physiological variations inherent in
batch culture systems obscured the relationship between algal growth
rate and external and internal phosphorus concentrations.
In continuous culture systems (chemostats), at steady state, all
environmental and physiological properties are constant. Then the
external phosphorus concentration is regulated by the specific growth
44
-------
rate of algae or the hydraulic residence time of the system (M-M, M
equation). In the case of excess phosphorus uptake the only parameters
that would be influenced are the cell concentration and the yield coef-
ficient. Logically, one can expect that at low residence times (low
cell age — high growth rate) in chemostats with no cell recycle, the
"excess" nutrient taken up by the cells is not completely utilized for
cellular growth before the cells leave the system. Therefore, the cell
yield coefficient for the specific nutrient is relatively low. However,
as the residence time (mean cell age) increases (i. e. , growth rate
decreases), increasing amounts of the "excess" taken up by the cells
are utilized for cellular growth resulting in higher yield coefficients.
Finally, when the mean cell age approaches infinity (0C = 0 -» °°)
essentially all of the "excess" nutrient taken up will be utilized for
cellular growth and the yield coefficient approaches its maximum
value, Ymax.
Nutrient Uptake Rate and Nutrient Utilization Rate. The nutrient up-
take rate in a chemostat can be defined as the amount of nutrient
taken up per unit cell mass per unit time, i. e. , the nutrient removal
velocity, q;
s - s
-L - - ° 1 mass nutrient removed ..
X dt ~ ^ Xj© cell mass-day ' '
From the cell continuity equation (Equation 20) the following simple
continuity or dimensional analysis can be obtained:
1 , . 1 dX dX 1 dS „ „ So " Si
M--0+kd-x dt - dS S "HE - Yq - Y Xle ^ *
ma8s of cells produced _ mass of cells produced mass of nutrient removed
cell mass-day ~ mass of nutrient removed cell mass-day
With a constant yield coefficient, Y, the cells utilize the limiting
nutrient for growth at the same rate at which the nutrient is taken up
from solution, hence no excess nutrient is accumulated in the cells
and the nutrient removal velocity, q, equals the nutrient utilization
rate, |a/Y (curve A, Figure 8). However, if "excess" nutrient uptake
occurs, the nutrient rerrjoval velocity, q, exceeds the utilization rate,
(curve B, Figure 8). Then,
=Yn (1U+ «CV>^ <33)
45
-------
EXCESS NUTRIENT REMOVAL
VELOCITY, q= -~
Nutrient utilization rate * nutrient removal
velocity; no excess uptake
when 9
SPECIFIC GROWTH RATE ,/t, time
-i
FIGURE 8. COMPARISON OF NUTRIENT REMOVAL VELOCITY AND
NUTRIENT UTILIZATION RATE UNDER NO "EXCESS"
AND "EXCESS" NUTRIENT UPTAKE CONDITIONS
The shaded area in Figure 8 represents the difference between the up-
take rate (with "excess" uptake) and the utilization rate under conditions
of no "excess" uptake between the limits of growth rate, |i = k
-------
Kinetics of "Excess" Nutrient Uptake in Chemostats. A flow diagram
definition sketch for the supply, uptake, and utilization of a specific
nutrient for growth in a chemostat at steady state is shown in Figure
9. The nutrient passes through the cell walls and membranes at the
nutrient removal rate, q, and is distributed to the functional processes
of the cell for cellular synthesis. The nutrient removal velocity, q,
in a chemostat is a function of cell age (residence time) and the yield
coefficient, i, e. ,
W-y na+Vdr-W- <*>
The nutrient utilization rate constant, kj, is a growth parameter
specific for the algal species, nutrient, and the environment in which
the "excess" nutrient taken up or stored is utilized for growth. In the
absence of any theoretical basis for the kinetic model of this reaction,
it is assumed to be a first order reaction according to:
dS -dS
HfT " ~d6^ ~ "k.Se <35>
where
S = nutrient concentration available for cell growth
including "excess" nutrient uptake and storage
S = nutrient concentration utilized by cells for their
® growth, and this amount = AqXj
A = nutrient content of individual cell at infinite cell age
in chemostat = (SQ-S )/Xmax
A = nutrient content of individual cell at cell age 0 in
chemostat = (Sq-SjJ/Xj.
In accordance with a materials balance for the specific nutrient in
the chemostat system, the following equation must hold:
nutrient
supply
extracellular nutrient
remaining in
chemostat
nutrient content of
individual cell
at 0 = ®
intracellular nutrient
available for
cell growth
steady state cell
concentration at
cell age 0
1
S =
o
Si + se + AoXi
(36)
47
-------
s,
A o = NUTRIENT CONTENT OF INDIVIDUAL CELL AT INFINITE CELL AGE
IN CHEMOSTAT , ^ " S>
^max
A = NUTRIENT CONTENT OF INDIVIDUAL CELL AT HYDRAULIC
RESIDENCE TIME,©,
xi
Se = NUTRIENT CONCENTRATION AVAILABLE FOR CELL GROWTH
INCLUDING EXCESS NUTRIENT UPTAKE AND STORAGE
*max MAXIMUM CELL CONCENTRATION IN CHEMOSTAT AT INFINITE
CELL AGE
Sg = NUTRIENT CONCENTRATION UTILIZED FOR GROWTH BY CELL
FIGURE 9. DEFINITION SKETCH FOR A SPECIFIC NUTRIENT
SUPPLY, UPTAKE AND UTILIZATION FOR GROWTH
IN A CHEMOSTAT AT STEADY STATE
The differential equation (Equation 35) can be integrated between two
limits; from 0 = 0, X! = 0, to 0 = 0, Xj = X3, with a discontinuity
when 0 £ 0W as shown in Figure 10. When 0 = 0W all the cells are
washed out from the chemostat
In S
S6 = In (SQ - Sj - AoXt)
e at
0 = 0
Xi =X!
X = 0
1
= -k,0
0 = 0
0 = 0
In
-------
>
N
2
a
tn
Q»
cn
tn
»
tn
z
o
DC
UJ
o
z
o
o
UJ
z
o
a.
2
O
o
0 ew
HYDRAULIC RESIDENCE TIME,©, OR MEAN CELL AGE,0c
FIGURE 10. COMPONENT NUTRIENT CONCENTRATIONS FOR
VARYING MEAN CELL AGES FOR A SPECIFIC
NUTRIENT IN STEADY STATE CHEMOSTAT
OPERATION WHEN "EXCESS" NUTRIENT UPTAKE
OCCURS
Since A = (S - S. )/X and substituting
O ixlcLX
'So-Si)Xi -kl6
=(S0-Sl)e . (37)
max
49
-------
Rearranging Equation 37, the following simple relation is obtained to
describe the cell concentration, in a chemostat in terms of the
maximum steady state cell concentration, Xmax, for a specific
nutrient supply to the chemostat.
It is apparent that the net yield coefficient, Yn = X}/(SQ -Sj), varies
exponentially with the hydraulic residence time (mean cell age) in a
chemostat without cell recycle.
(39)
where
Y = maximum cell yield coefficient = X / (S - S,)
max max o 1
Ka = nutrient utilization constant corresponding to
the reciprocal of the nutrient utilization rate,
i. e. , = 1/kj,
The distribution of the nutrient components, Sj, Se, and Sg, as a
function of hydraulic residence time, 0, (no cell recycle) or mean
cell age, 0C, in a chemostat is shown in Figure 10.
Empirical Growth—Yield Function. Fuhs [96] proposed empirically
that for two marine diatoms, Cyclotella nana and Thalassiosira
fluvialilis, the relative specific growth rate, |j./{a, is related to the
relative amount of phosphate per cell, A/A0, as follows:
/ (1 - A/A ) \
£ = [i - 2 ° ) (40)
where
A = amount of phosphorus per cell
Aq = minimum amount of phosphorus per cell.
In terms of the cell yield coefficient, Fuhs' equation can be re-
arranged as,
Y„ = Ymax (ln'2 -til - n/A)) ' ^
50
-------
The equation of Fuhs (Equation 40) empirically describes the relation-
ship between |i and intracellular phosphorus concentration for two
marine diatoms. On the other hand, Equation 39 is proposed for the
general description of the phenomenon of "excess" nutrient uptake in
chemostats in terms of a varying yield coefficient which is a function
of mean cell age.
Rectangular Hyperbola Yield Function. Because of the general
acceptance of the rectangular hyperbolic functions (e. g. , the M-M, M
model) for relating growth rate and limiting-nutrient concentration,
a rectangular hyperbolic function appears to be logical for the relation
between the cell yield coefficient and mean cell age as follows:
:)
Y«-Vm«(5^Tql (42)
where
Kg = a nutrient utilization constant corresponding to
the mean cell age at Yn = Ymax/2.
A graphical comparison of the exponential function (Equation 39) and
the rectangular hyperbolic function (Equation 42) for a specific value
°f Ymax is shown in Figure 11,
Fraction of "Excess" Nutrient Uptake, fe. The amount of "excess"
nutrient uptake, Se, can be described based on Equations 36 and 39
follows:
Se = So " S. " AoX, =
max
"ec/ICA
= (S0 - S,) I° J ¦ (44)
Equation 44 indicates that the amount of "excess'1 nutrient uptake is
proportional to the amount of total nutrient uptake, SQ - Sj, and varies
exponentially with the hydraulic residence time. The fraction of
"excess" nutrient uptake, fe, defined as the fraction of the total
nutrient uptake which is accumulated and/or stored inside the cell
(not yet utilized for growth), Se/ (SQ - Sj), is a simple function of mean
cell age or hydraulic residence time. This can be shown as follows:
51
-------
MEAN CELL AGE , 0C
FIGURE 11. GRAPHICAL COMPARISON OF EXPONENTIAL
FUNCTION AND RECTANGULAR HYPERBOLIC
FUNCTION FOR NET YIELD COEFFICIENT -
CELL AGE RELATIONSHIP (SAME VALUE OF
MAXIMUM CELL YIELD, YMAX)
As an exponential function based upon Equation 39.
Y -0 /Ka
fe = S~rs7 = 1 - 5T-1- = e ° : (45)
o i max
52
-------
As a rectangular hyperbola based upon Equations 42 and 43,
f 2
e S - S.
o
S
e
(46)
Equations 45 and 46 are shown graphically in Figure 12.
Chemostat Kinetic Equations with a Varying Yield Coefficient. Kinetic
equations for the chemostat with a varying yield coefficient caused by
"excess" nutrient uptake can be summarized as follows:
Exponential Model. Cell continuity equation:
(27)
where
S - S.
o
(25)
Cell yield coefficient - mean cell age equation;
(39)
where
Y
53
-------
HYDRAULIC RESIDENCE TIME, 6, OR MEAN CELL AGE, 0C
FIGURE 12. THE FRACTION OF EXCESS NUTRIENT UPTAKE
AS A FUNCTION OF MEAN CELL AGE
(COMPARISON OF EXPONENTIAL AND
HYPERBOLIC MODELS)
Specific growth rate limiting nutrient concentration relationship;
where
(^7+^)
M- - I if "1 a I (I*)
^ + kd (20)
£ =W~ +kd' <30>
w
54
-------
Steady state cell concentration;
( "VKa\
X. = Ymax ^ e J ¦
(47)
Rectangular Hyperbolic Model. Cell continuity equation:
=Y.q-kd . (27)
c
Cell yield coefficient - mean cell age equation
Y 0
V max c /a?)
n ~ 0 + Kn
c o
where
Y - X1
n S - S
o i
X
Y max
max S - S '
o 1
Specific growth rate limiting nutrient concentration relationship is:
^ = ^ ^ Kg + Sj 1
where
H = | + kd (20)
^ = F" + kd ' (30)
c
Steady state cell concentration is:
= p0 -s.' >rK; • f»»)
C B
55
-------
Evaluation of Chemostat Growth Kinetic Constants, The general
bio stimulatory growth parameters of major concern in the chemostat
are: steady state cell concentration, Xjj maximum specific growth
rate, ji; the half saturation constant, Kg; the nutrient utilization
constants, and Kg; and the decay rate of the organism, k
-------
>
2
Z
o
5
oc
UJ
o
z
o
o
UJ
a
UJ
I
>
Q
<1
UJ
f—
tn
HYDRAULIC RESIDENCE TIME,©, OR MEAN CELL A6E,9C
FIGURE 13. THEORETICAL STEADY STATE CONDITIONS IN
CHEMOSTAT WITH A VARYING YIELD
COEFFICIENT (EXCESS NUTRIENT UPTAKE)
Nutrient Utilization Constanta, and Kg. The nutrient utilization
constant, Ka> is the mean cell age at which the net yield coefficient
(Y„) is equal to 0. 632 maximum yield coefficient, i. e. , Yn =
0. 632 Ymax in the exponential yield model. Similarly, the nutrient
utilization constant, Kn, is the mean cell age at which the net yield
coefficient is one-half the maximum yield coefficient in the rectangular
hyperbolic model (Figure 11). The nutrient utilization constants
represent the rate of nutrient utilization for growth relative to the
nutrient removal rate for a specific nutrient, algal species, and
57
-------
environmental conditions. The higher the value of nutrient utilization
constants (K^ or Kg), the lower the nutrient utilization rate (e. g. ,
kj = 1/Ka) and an increasing amount of "removed" nutrient is
accumulated and stored inside the cell, "With no "excess" nutrient
uptake, the nutrient utilization constants will be zero and the net
yield coefficient approaches a constant value equal to Ymax.
Decay Rate, k^. The decay constant, k
-------
SECTION V
METHODS AND MATERIALS
ORGANISM SELECTION AND RELATED FACTORS
Algal Species
In selection of the algal species to be used in algal bioassays, con-
sideration was given in the Provisional Algal Assay Procedures [35]
to the characteristics required from such an organism. It was recognized
that the algae should have a broad nutrient response, a distinct shape,
and a uniform size. It was further deemed necessary that it should
divide distinctly, not attach to glass or surfaces, stay in suspension
with slight agitation, not clump, grow at the maximum rate in a medium
simple to constitute, be nonauxotrophic, not excrete toxic substances,
and normally be associated with oligotrophic waters. These consid-
erations led to the selection of Selenastrum caprLcornutum Printz as a
test organism. This alga has cells which are characteristically lunate
with apices acutely pointed; cells are approximately 2 n wide and 20 |j.
long and it occurs in both oligotrophic and eutrophlc waters [48]. The
distinct morphology, little form variation, the fact that cells occur
mostly singly and stay in free suspension, and its obligate autotrophic
nature make this organism a superior test alga. For these reasons
cultures of Selenastrum capricornutum were selected for use in the
work described in this report. These cultures provided initially by
O. Skulberg (NIVA, Oslo, Norway) were obtained from a common
stock culture maintained by T. E. Maloney of the Pacific Northwest
Water Laboratory, Corvallis, Oregon. The culture of S. capricornutum
was plated several times to eliminate possible algal contaminants and
to minimize bacterial contamination.
Nutrient Medium
The nutrient medium outlined in the Provisional Algal Assay Procedures
[ 35] was used as basic nutrient medium. The medium is designated
100% PAAP and its macronutrients are listed in Table I and the micro-
nutrients in Table II. The reagents used were ACS reagent grade
chemicals.
Inoculum Preparation
A 7-day old culture of S. capricornutum was centrifuged and the
supernatant discarded."" The sedimented cells were resuspended m
distilled water so that when this suspension was added to the test flasks
it resulted in the required inoculation concentration of 1000 cells/rru-
59
-------
TABLE I
MACRONUTRIENT COMPOSITION OF 100% PAAP MEDIUM
a
Components
Concentration
mg/.?
Essential
Nutrient
Concentration
mg/l
NaN03
85. 00
N
14. 00
k2 hpo4
3. 48
P
0. 62
MgCl2
19. 00
Mg
9- 68
MgS04 • 7H2 O
49. 00
S
6. 37
CaCl2" 2HzO
14. 70
Ca
4. 01
Na2 COj
50. 00
K
1. 56
FeCl3
0. 32
Fe
0. 11
Na2 EDTA* 2HzO
1. 00
SI
Before addition of iron and trace metals the pH was adjusted to 7. 0
to minimize precipitation.
Na2EDTA = Disodium ethylene diamine tetra acetic acid.
TABLE II
MICRONUTRIENT COMPOSITION OF 100% PAAP MEDIUM
Component
Concentration
M-g/^
Essential
Nutrient
Concentration
h3 bo3
618. 40
B
110. 00
MnCl2
880. 88
Mn
380.00
ZnCl2
109. 03
Zn
50. 00
CoCl2
2. 60
Co
1. 18
CuCl2
0. 03
Cu
0. 01
Na2Mo04' 2H2 O
24. 20
Mo
9- 60
Na2EDTA- 2HZ O
7440.
Note: The trace metals and EDTA were combined in a single stock
mix at a level of 1000 times the final concentration.
60
-------
Culture Environment
Both batch and chemostat experiments were conducted in a constant
temperature room (24 °C ± i °C) equipped with "cool white" fluorescent
tube lighting to provide a light intensity of 400 ft-c (4300 lux) ±10%.
Light intensity for the chemostats was measured adjacent to the reactor
at the top, middle, and bottom of the liquid column at points on the
reactor nearest to the light source. For batch tests, the light intensity
was measured adjacent to the flask at the liquid level.
The pH was maintained in the chemostats between 7. 0 and 8. 5 by aerating
from the bottom with air or an air-COz mixture. The COz content of the
air was not maintained at a fixed concentration but was varied as nec-
essary to maintain the desired pH range. The pH in batch tests was
maintained in the same range (7. 0 to 8. 5) by ventilation with air or
air-COz mixtures.
WATER SAMPLES
Four different water samples were used in this study. Samples were
collected when needed and were not stored for long periods before
sample pretreatment was carried out.
Clear Lake Water
The Clear Lake water samples were taken from the pump inlet at the
Clear Lake Mosquito Abatement Research Laboratory. The inlet was
located about 50 yd offshore in an area where no septic tank outlets or
other sources of waste discharges were located.
Sacramento River Water
The samples of Sacramento River water were taken off the river bank
about a half mile upstream of the bridge at Knights Landing at the
California Department of Water Resources sampling station.
Lake Anza Water
The Lake Anza water samples were taken at the outflow weir in Tilden
Regional Park, Berkeley.
Secondary Effluent
Secondary effluent samples from a pilot plant treating domestic sewage
from the City of Richmond at the Sanitary Engineering Research
Laboratory, University of California were used.
61
-------
SAMPLE PRETREATMENT METHODS
Several different sample pretreatment methods were evaluated in this
investigation.
Auto claving
The water samples were autoclaved at 121 °C for 15 min, cooled down
to approximately 25 "C, and used in assays.
Pasteur izat ion
The water samples were pasteurized by heating to ?i°C (in some
cases boiling point) for approximately 10 sec followed by cooling down
to 25°C before use.
Filtration
Depending on the assay, samples were either filtered through Whatman
No. 41 filter paper or GF/C glass fiber filters or through Millipore
membrane (0. 45 (j.) filters. If necessary, pressure was used to improve
filtration rates.
ANALYTICAL METHODS
Biomass Analyses
Biomass was determined by the following four methods:
Absorbance. Absorbance was measured against a distilled water blank
at 600 mjj. in a one inch cell in a B and L Spectrometer 20. Results
were expressed as OD/inch.
Suspended Solids. The determination of suspended solids was made by
use of preweighed glass fiber filters (GF/C) which had been previously
rinsed with distilled water. Samples of known volume were filtered,
the filters dried at 105 °C and then placed in individual desiccator jars.
Suspended solids in mg/f were determined by difference after 24 hr.
Cell Counts and Total Cell Volume. Cell counts were determined with
a Coulter counter Model B and total cell volume estimates were obtained
using a mean cell volume computer (Coulter Electronics). Results
were expressed as cell number/mi and x 106/mi, respectively.
62
-------
Nutrient Analyses
Nitrates and Nitrites (Soluble). Nitrates were determined using the
cadmium reduction technique according to Strickland and Parsons [ 105],
The nitrate is first reduced to nitrite when the sample is run through
a column containing amalgamated cadmium fillings. The nitrite thus
produced plus about 95% of the original nitrite present in the sample
is determined by diazotizing with sulphanilamide and then coupling with
N - (1-naphthyl) ethylendiamine to form a highly colored azo dye, the
extinction of which is measured at the wavelength of 543 m|j,.
Organic Nitrogen. Organic nitrogen was determined by distillation and
Kjeldahl techniques according to Standard Methods [ 106].
Ammonia. Ammonia was determined with the colorimetric method of
Solorzano [107] measuring the blue color of indophenol after the reaction
at high pH of ammonia, phenol and hypochlorite.
Orthophosphates (Soluble). Orthophosphates were determined according
to Strickland and Parsons [ 105]. The sample after filtration is
allowed to react with a composite reagent containing molybdic acid,
ascorbic acid, and trivalent antimony. The resulting complex heter-
opolyacid is reduced m situ to give a blue color, the extinction of which
is measured at 885 m^i.
Total Phosphorus. Before determination of total phosphorus an appro-
priate volume (10 - 40 mi) of the unflltered sample was digested with
9 mi of HNO3 and 1 mi of H2S04 in 100 mi Kjeldahl flasks until the
white fumes of NOz gas disappeared. The volume of sample remaining
was diluted and the pH value adjusted to about 4 with Na2C03 . After
the sample volume was brought up to 100 mi, the determination was
made as for orthophosphates.
Precision and Accuracy of Analyses. The precision and accuracy of
analyses were given by Porcella et al. [ 34] except that the coefficient
of variation of orthophosphate analyses at extremely low concentrations
was reduced by exerting extreme caution when conducting the analyses.
The coefficient of variation for cell counts and cell volume estimates
by use of an electronic particle counter was also determined. The
values of these coefficients for the orthophosphates, cell counts, and
cell volumes, as well as those previously reported, are summarized
in Table III.
63
-------
TABLE IH
ACCURACY AND PRECISION OF ANALYTICAL METHODS
AND BIO MASS ESTIMATES
Analysis
Mean
Coeffic ient of
Variation3, (%)
pH
7, 98
0. 5
Absorbance
0. 026
5. 1
OD/one-inch cell
0. 113
1. 9
0. 33
0. 9
Hemacytometer cell counts
823
14. 8
Cells/mm3
1420
9. 9
2010
8. 3
Coulter counter cell counts
17. 3
3.4
Cells x 106 /ini
118. 3
1. 6
Total cell volume
1. 6
3. 2
H3 x 106 /ml
12. 7
i. 3
Suspended solids
9.41
7. 8
mg/i
16. 5
1. 3
74. 0
3. 1
Suspended carbon
4. 68
13. 3
mg/£
17. 7
1. 7
58, 7
1. 1
no3
10. 7
13. 5
[ig/Jt as N
1520
4, 1
3970
2. 2
P04
0. 7
20. 9
jxg/i as P
5. 1
7. 6
75. 5
4. 8
303
1.4
Total P
135
14. 0
Hg/i as P
561
2. 0
840
6. 9
Coefficient of variation (%) = 100 (standard deviation)
(mean *)•
64
-------
ASSAY TECHNIQUES
Bottle or Batch Culture Assays
Pyrex glass 500 mi Erlenmeyer flasks were used for batch culturing. .
The flasks were cleaned carefully according to the Provisional Algal
Assay Procedures [ 35], and 250 mi quantities of either media or
natural water samples were transferred to individual flasks. The media
or water were inoculated with Selenastrum capricornutum and the flasks
were closed with 150-mi Pyrex beakers (inverted) and connected to
the ventilation system. This ventilation system could utilize either air
or an air-COz mixture (COz in concentrations as needed). The gas
was filtered through glass wool and activated carbon and was then
passed through an acid solution and distilled water successively before
being distributed to each flask as shown in Figure 14 (see also Appendix
A). The rate of gas flow to each chemostat or to the flasks was reg-
ulated through in-line rotameters. The gas flow was regulated to
maintain the pH in the flasks between 7. 0 and 8. 5. The flasks connected
to the ventilation system were then clamped onto a rotary shaker (New
Brunswick Scientific Co. , Model VS) and shaken at 120 oscillations/min
as shown in Figure 15.
In general, an initial cell count and cell volume determination were
made on each flask immediately following inoculation. Thereafter
counts and volume estimates were made daily from the second to the
eighth day of incubation to allow estimates of the maximum specific
growth rate, batch (Jl^) of the algae. When the stationary phase was
reached in the cultures (14 to 21 days) a terminal cell count, cell
volume, and suspended solids determination were made on each flask.
Suspended solids measurements were not used to determine ji, because
the values are obtained early in an assay (first to third day), at
which time the suspended solids determinations are extremely imprecise
due to low values (<5 mg/i). In addition to the above, the initial and
terminal nutrient concentrations in the natural samples were determined
to allow estimates of yield coefficients.
Continuous Culture (Chemostat) Assays
Chemostat Design. A schematic diagram of the chemostat algal culti-
vation system is shown In Figure 16. The chemostat reactor was
made of Pyrex glass with flat glass bottom, the top closed by a "cafe-
au-lait" stopper, a siphon break was provided in the Input line, and it
was provided with magnetic stirring. In the feeding system, glass
tubing was used whenever possible; minimum lengths of Tygon tubing
were used for joints. Each chemostat was aerated with an air-C02
mixture at an air flow rate of approximately 0. 5 ft3 /hr; the specific
flow rate depending on the amount necessary to maintain the pH in the
65
-------
1 AIR SUPPLY
2 REGULATORS
3 GLASS WOOL
4 ACTIVATED CARBON
O.I N H2S04
6 GLASS BEAD TRAP
7 DISTILLED H20
8 ROTAMETER
9 C02 SUPPLY
O MANOMETER
II TO GROWTH UNITS
FIGURE 14. AIR TREATMENT AND DISTRIBUTION SYSTEM
66
-------
//
FLUORESCENT LIGHTS
MANIFOLD WITH
CAPILLARY TUBES
AIR FLOWMETER
FILTERED AIR INLET
VENTILATED CULTURE
FLASK
SPRING CLAMPS
—Y ^— PLATFORM FOR
ROTARY SHAKER
FIGURE 15. BATCH ASSAY UNIT
67
-------
12
XE
o o
tpl
£
10
1.
FEED BOTTLE
7
MAGNETIC STIRRER
2
POLYSTALTIC PUMP
8
SAMPLING POINT
3
SIPHON BREAK
9
AIR FLOW RATE CONTROL
4
CHEMOSTAT REACTOR
10
COTTON WOOL AIR FILTER
5
FLUORESCENT LIGHT
II
ROTAMETER
6
EFFLUENT BOTTLE
12
AIR-C02 INPUT
FIGURE 16. DIAGRAM OF CHEMOSTAT ASSAY UNIT
68
-------
cultures between 7, 0 and 8. 3, The pH can be controlled by altering
the air flow rate (within the range of 0. 1 to 1.0 ft3/hr) or by altering
the amount of C02 in the air. COz was added as steady state was
approached and thereafter during steady state only aeration at 0. 5 ft3/hr
was used. The air was filtered through an acid and water wash after
passing through activated carbon and glass fiber columns (Figure 14),
Chemostat Operation. The chemostats, including all tubing and the
medium, were autoclaved at 121 °C for 15 min. The chemostats were
then filled with the sterilized medium, inoculated, and the cultures
allowed to grow before a constant feed rate was established. The
chemostats were operated at a constant feed rate for at least two
hydraulic residence times. After this period the cultures were operated
until "steady state" conditions were established.
After the cell mass concentration in the chemostat had reached its
initial maximum value, the attainment of a near constant cell mass con-
centration implied that steady state had been reached, "Steady state"
for this investigation was defined as a period of at least one residence
time wherein the cell mass concentrations, determined at daily intervals,
did not vary more than ± 20% from a mean value- This variation was
to be random, i. e. , the cell mass concentration could not exhibit a
consistent upward or downward trend. In general the tests were con-
tinued for 3 or more residence times before the steady state values
were obtained.
To minimize volume changes inside the chemostats, only 20-mi samples
were withdrawn daily from the reactors. Analyses of pH, absorbance,
and cell counts were made on the reactor samples. All steady state
analyses of biomass (absorbance, suspended solids, and cell counts)
and nutrients were conducted on 100-mi samples withdrawn from
chemostats when steady state was reached. The volume of the effluents
were collected and measured daily to ensure constant residence times
during the entire assay period. The complete analyses scheme is
shown in Figure 17.
Computational Methods. A computer linear least square regression
analysis was used to estimate the kinetic growth constants for all
linear kinetic equations and all linear transformations from nonlinear
equations. A nonlinear computer program (COMMON FILE CAL G2
NLIN, Appendix B) was also used to estimate the constants directly
from nonlinear regression analyses.
69
-------
DAILY MEASUREMENTS
/
Effluent
Volume
20 ml sample
from chemostot
I
Jl
pH, absorbonce,
cell counts
Filtrates
*
'
i
9
\
Nitrite
Nitrogen
Nitrate
Nitrogen
Ammonia
Nitrogen
Phosphate
Phosphorus
STEADY STATE MEASUREMENTS
L
100ml sample collected
from chemostot
20 ml
Total
Phosphorus
80 ml
Filter through fared
GF/C Filters
'
Dried anc
Suspende
weighed
d solids
FIGURE 17. FLOW DIAGRAM OF ROUTINE ANALYSES
PERFORMED ON CHEMOSTATS
70
-------
SECTION VI
RESULTS AND DISCUSSION
BATCH CULTURE STUDIES
Sample Pretreatment Evaluation
In accordance with a decision of the PAAP evaluation group meeting
in Corvallis, Oregon in May 1970 [ 108], four different sample pre-
treatment methods, i. e. , autoclaving, pasteurization, and two
filtration techniques were evaluated. Each treatment was evaluated
for its 1) ability to remove indigenous algae from samples, 2) effect
on the nutrient concentrations in the sample, and 3) effect on the
algal response in batch assays.
Removal of Indigenous Algae. To simulate a natural water, an' aliquot
from a high-rate algal pond was spiked into 100% PAAP medium
(Tables I and II) in a proportion of 1 to 99, respectively. Samples
(250 mf) of this mixture were treated by different pasteurization and
filtration techniques (100°C for 10 sec, 71°C for 15 sec). The samples
were then incubated under standard PAAP conditions [3 5] while
ventilating the cultures as described by Porcella et_aL_ [34]. The
absorbance of the cultures were determined initially and at termination
12 days later.
The absorbance values (Table IV) showed that although the control
flasks had a tenfold increase in absorbance, none of the pasteurized
or filtered samples showed growth. Therefore, it may be concluded
that algae were removed effectively from the PAAP medium by
pasteurization and filtration techniques.
To determine if indigenous algae of a natural water sample would also
be removed by the pretreatment methods, a grab sample from Lake
Anza, (Berkeley, California) was used. This sample was halved and
one-half was spiked with high-rate algal pond water (1 to 99 ratio) to
ensure the presence of algae. Both the spiked and the nonspiked
waters were divided into 4 aliquots and subjected to the pretreatment
methods, i. e, , autoclaving, pasteurization (71"C for 15 sec), and
filtration through Whatman No. 41 and membrane (0. 45-[j, pore size)
filters. The resulting waters were transferred asBptically to batch
assay flasks, using three replicate flasks for each treatment. The
flasks were incubated (without inoculation) under standard PAAP
conditions [35"] for three weeks. Two cell counts, using an electronic
particle counter, were performed during incubation (12th and 21st
days).
71
-------
TABLE IV
REMOVAL OF ALGAE FROM PAAP MEDIUM BY
PASTEURIZATION AND FILTRATION
Treatment
Initial
Absorbance
Terminal
Absorbance
(12 days)
Observation
Pasteurization
710 C for 15 sec
0, 015
0. 015
No growth
Pasteurization
100°C for 10 sec
0. 015
0. 015
No growth
Filtration
Whatman No. 41
0. 015
1
! o
o
: -
!
No growth
Filtration
Membrane, 0. 45fj.
0. 015
0. 015
No growth
Control
0. 015
0. 16
Growth
The results are summarized in Table V. An untreated control
indicated that the maximum count obtainable in the water was approxi-
mately 145 x 103 particles/m?. Except for Whatman filtration, none
of the »ther procedures approached this value. However, interpreta-
tion of the results is difficult due to a high initial background count in
the samples (approximately to 60 x 103 particles/mi)- The lower
values for the heat-treated samples suggest that these treatments
removed the indigenous algae; however, the values for membrane
filtration samples indicate that this method was only partially effective
in removing the indigenous algae, Whatman filtration treated samples
showed the highest values, hence Whatman filtration does not appear
to be effective in removing indigenous algae, It is noteworthy that
membrane filtration through 0, 45jj. membranes was found by Dixon
[109] to be unable to remove a small alga from marine samples.
The high background counts obtained with the electronic particle
counter is probably due to debris normally associated with natural
water eamples. Because the electronic counter does not distinguish
between debris and algal cells, such counts are not reported as alga>l
counts but rather as particle counts. This shortcoming limits the ' 1
usefulness of these instruments, particularly in assays of low-nutrient
concentration natural samples. Under these circumstances, micro-
scopical observations to verify the presence 6r absence of debris is
necessary.
72
-------
TABLE V
REMOVAL OF INDIGENOUS ALGAE FROM A
NATURAL SAMPLE BY PRETREATMENTS
Pretreatment
Method
Aigai
Spiking
Counts x 1000/m?k
12th day
21st day
Pasteurization
71°C for 15 sec
Not spiked
42., 3 (85%)c
32. 5 (40%)
Spiked
69. 6 (58%)
34. 1 (2%)
Autoclaving
Not spiked
19. 0 (123%)
34. 7 (154%)
Spiked
11. 5 (39%)
14., 3 (95%)
Filtration
Membrane, 0, 45fi
Not spiked
41. 2 (78%)
107. 0 (65%)
Spiked
40. 0 (138%)
56. 6 (103%)
Filtration
Whatman No, 41
Not spiked
62, 5 (13%)
101. 9 (25%)
Spiked
86. 8 (16%)
162. 9 (19%)
Control
Not spiked
146. 2
145. 9
3.
One part high-rate algal pond water to 99 parts sample.
Mean of 3 replicates.
Q
Value in parentheses = coefficient of variation = standard
deviation (100)/mean.
Pretreatment Effect on Chemical Composition of Samples. The effect
of pretreatment methods on the chemical composition of samples was
examined in three separate experiments on grab samples taken from
Lake Anza, These grab samples were taken at different time intervals
and were not identical in composition. Each of the samples was
divided into four equal parts- Each part was treated with one of the
four sample pretreatment methods and each chemical analysis was
made in triplicate.
The results of the chemical evaluations are summarized in Table VI.
Sample pretreatment apparently had no significant effect on the
chemical composition of the samples because no consistent trends
were observed. For instance, while the NH3-N content was reduced
by membrane filtration in two experiments, it was increased in the
73
-------
TABLE VI
PRETREATMENT EFFECT ON THE CHEMICAL COMPOSITION OF SAMPLES
Sample Pretreatment j
Chemical
Analyses
Experiments
Autoclaved
Pasteurized
Whatman
No. 41
Filtered
Membrane
Filtered
Untreated
Sample
PO4-P
fig/'
I
II
III
8. la
6. 1
4. 1
8. 2
6. 3
4. 0
7. 6
6, 6
3. 5
6. 1
5. 8
2. 4
7. 2
4. 8
2. 1
NH3-N
H-g/*
1
II
III
173
45. 5
64.4
143
55. 2
88. 9
228
71. 5
70. 0
232
24
16. 1
153
68. 3
49. 0
N03 and
NOz -n
^g/'
I
II
III
8. 6
2. 7
0. 9
10. 9
4. 6
11. 5
12. 5
6. 2
13. 2
25
13. 6
17. 3
10. 6
4. 7
15. 3
Total-P
^g/'
I
II
HI
198
105
103
153
138
126
130
122
66, 8
not det.
not det.
not det.
3 93
128
110
Total-N
v-gti
I
H
147
975
304
592
216
957
not det.
not det.
231
665
Each value represents mean of three replicate determinations.
-------
third experiment. The same phenomena were observed for the other
constituents assayed. The absence of clear trends in these analyses
is probably due to the low precision of the analytical techniques at
the extremely low concentrations of chemicals (all less than 1 mg/f
and mostly less than 100 p-g/i) encountered.
Therefore, it is conceivable that the filtration techniques, especially
membrane filtration, removed substantial quantities of suspended
materials (evidenced by visible green layers on filter pads after
filtration), resulting in substantial losses in total nitrogen and
phosphorus from solution. Unfortunately total nitrogen and phosphorus
were not determined for the membrane filtered samples because of
lost samples. Because only a small fraction of the total nitrogen and
phosphorus occurred in dissolved inorganic form (Table VI), such
losses may result in the reduction of the maximum cell production
responses in algal assays.
Effect of Pretreatment ^ Responses. To evaluate the effect of
pretratment on the algal responses of an assay o e Each
a grab sample of Lake Anza was divided into four portions Each
portion was treated with one of the pretr*"toent me&od.. i^e ,
pasteurization (71°C for 15 sec), autoc laving fil tr ation ttirough
either Whatman No. 41 or membrane (0. 45-n) filters. Jkereafte
four replicate batch flasks (250-m« sample in each) were Pr^reJ
for each pretreatment. The flasks were in?^JaA?>
capricornutum and incubated under standard PA ^ ^ co ^
with air ventilation. Cell counts were performed ^"lj Jjjf
to the 8th day. Terminal suspended solids were too low for accurate
measurement and therefore final cell counts were made on the 14th
day.
Mean Snecific Growth Rate <&). A comparison of the algal growth
rates Used on cell number. Obtained daring the.batch assays »
given in Table VII. The highest growth rates were obtained in
the autoclaved samples, followed by the pasteurized samples.
The growth rates obtained with the filtration treated samp Is
were the lowest. The growth rates m the autociaved sample8
were significantly higher ft, = 0. 05) than the growthratea of
either filtered treatments. The pa,iteurued
differ significantly from the autoclaved or filtered samples,
however, the variation with this treatment was the greatest.
There was no significant difference between the filtered samples.
75
-------
TABLE VII
MEAN ALGAL GROWTH RATES OBTAINED
AFTER SAMPLE PRETREATMENT
Treatment
Mean Maximum
Specific Growth Ratea
day-1
Coefficient of
Variation
%
Autoclaved
0.46
26. 4
Pasteurized
0. 33
43. 0
Filtration Whatman
No. 41
0. 25
24. 0
Filtration Membrane
0.45-n
0. 25
29. 3
aMean growth rate of four replicate flasks; based on cell
numbers.
^Coefficient of variation = standard deviation (100)/mean.
A
Maximum Cell Concentration, X. A comparison of the maximum
cell concentration values for the different treatments is given in
Table VIII. The autoclaved samples had the highest mean value
but also the greatest amount of scatter (coefficient of variation =
80%). Similarly, the Whatman filtered samples had a mean of
more than iOO, 000 cells/m? and a wide scatter (Cv = 47%). These
means did not differ significantly from the mean of the membrane
filtered samples. However, the pasteurized sample mean (84 x
103 cells/m£), while less than those of the autoclaved and Whatman
filtered samples, differed significantly (t-test with p = 0. 05) from
the mean of the membrane filtered sample. The reason for this
was that there was less scatter in pasteurized replicates (Cv =
29%). If the autoclaved and Whatman filtered samples had less
variation they also would have differed significantly from the
membrane filtered samples. The autoclaved, pasteurized, and
Whatman filtered samples did not differ significantly from each
other.
76
-------
TABLE VIII
MEAN CELL PRODUCTION (X) VALUES OBTAINED
AFTER SAMPLE PRETREATMENT
Sample Pretreatment
A
X
Mean Maximum Cell
Production (14 days)
cells/mi!a
cv
Coefficient of
Variation*3
%
Autoclaved
113,000
80
Pasteurized
84, 200
29
Filtration through
Whatman No. 41
103, 000
47
Filtration through
membrane filter
33, 300
14
aMean of four replicate flasks.
Coefficient of variation (%) = standard deviation (100)/mean.
Williams and Yeiseley [59] in a similar evaluation of sample
pretreatment methods reported a 20-fold increase in maximum
cell production of autoclaved samples compared to filtered
(0. 5-(jtpore size).control. Therefore, it can be concluded that
heat treatment of samples would result in a higher maximum
cell production than samples treated by membrane filtration.
Summary of Sample Pretreatment Evaluation. All sample treatment
methods appeared to remove indigenous algae from the samples;
however Whatman filtration is likely the least effective. Autoclaving,
pasteurization, and membrane filtration appear to be adequate in
removing most indigenous algae.
The effects of the pretreatments on the chemical composition of the
water samples seemed to indicate no great chemical changes taking
place. However, filtration of samples, especially membrane filtra-
tion, which removes algae and particulate matter, must also remove
some of the suspended nutrient constituents of the waters. This effect
was clearly demonstrated by significantly lower algal growth rates and
cell concentration values for filtered samples when compared to
pasteurized samples and likely autoclaved samples.
77
-------
Two problems are posed by these results, namely: 1) Should nutrients
such as nitrogen and phosphorus which have become part of the biomass
(algal cells) or in particulate form in a water be removed prior to
assaying the water? Removal of these nutrients clearly results in
lower cell production values. Thus, the capacity of the water to
stimulate algal growth may be underestimated; 2) When algal growth
rates are to be measured, heat treatments such as autoclaving and
pasteurization will kill indigenous organisms and/or soluble particulates
resulting in a release of nutrients back into the medium thereby affecting
the algal assay growth rates. In the natural water these nutrients are
bound and may not be available for algal growth. Hence a lower algal
growth rate would be obtained and the assay of the heat treated samples
will tend to overestimate the in situ algal growth rate.
Therefore, when the potential of a water to produce algal cells is
determined no portion of the nutrients should be removed because
these nutrients form part of the total nutrient supply. In these cases
filtration techniques should not be used. When algal growth rates are
to be measured the release of bound nutrients back into the sample,
thereby affecting the growth rates, is to be avoided. Therefore, of
the methods evaluated the membrane filtration technique appears to
be the method of choice when growth rates are to be determined.
Interlaboratory Precision Tests
In accordance with the decision of the PAAP evaluation group meeting
at Corvallis, Oregon in May 1970 [ 108], studies were conducted as
part of the interlaboratory precision test for the batch culture assay.
A protocol for this test was supplied by Dr. C, E. Weiss of the
University of North Carolina. This protocol called for the use of a
single stock culture of Selenastrum capricornutum (supplied by
T. E. Maloney, Pacific Northwest Water Laboratory, Corvallis) and
a single source of chemicals (supplied by C, E. Weiss, University of
North Carolina). The participating laboratories were to determine
the algal responses (growth rate and cell production), at specified
values of pH, light intensity, and temperature, and at specific intervals,
i.e., 3, 5, 7, 9, 13, 17, and 21 days. Different concentrations, i.e.,
1%, 10%, 30%, and 100% of PAAP medium (Tables I and II) were used
and each dilution was assayed in five replicate flasks. Standard PAAP
conditions [35] were used and the use of foam plug enclosures was
specified,.
First Interlaboratory Precision Test, The protocol as outlined above
was followed although the use of foam plugs deviated from the standard
procedures [34] used in this laboratory. The results of the cell counts,
suspended solids, total cell volume, and pH measurements for the
first interlaboratory precision test are summarized in Table C-I,
Appendix C. The temperature values did not vary outside the range
78
-------
24 °C ± 2°C. The mean light intensity on the shaker table was 412 ft-c
(coefficient of variation = 4. 0%) as measured by a Weston (Model 756)
light meter. Temperature and light conditions agreed well with those
specified for the standard PAAP procedures [35].
The cell counts in the different dilutions of PAAP medium indicated
the virtual absence of growth in the 30%, 10%, and 1% concentrations.
In the 100% concentration only four flasks showed appreciable growth.
The fifth flask contained only 43, 000 cells/mI at termination compared
to more than 3 x 106 cell/m.? for the other four flasks. It was suspected
that the foam plug enclosures were toxic to Selenastrum capricornutum
and only in the higher nutrient concentrations could the toxicity be
overcome by the alga. This conclusion was verified by results of
Scherfig et_al^ [76] which showed that some kinds of foam plugs are
extremely toxic to Selenastrum capricornutum, and foam plug toxicity
was also observed by Payne [60] and Weiss [68],
The use of foam plugs in batch culture assays cannot be recommended
because of this apparent toxicity. Observations that in one batch of
foam plugs some are toxic and some not, indicate that all foam plugs
must be tested for toxicity before being used in algal assays [77],
This is a cumbersome procedure and it does not remove all concern
about possible toxic effects. Hence the use of foam plugs should be
avoided whenever possible.
Second Interlaboratory Precision Test. The interlaboratory precision
test was repeated because of the foam plug toxicity experienced in the
first test. Instead of foam plugs, ventilated flasks with glass beaker
tops (Figure 15) were used. The same flasks used in the first precision
test were used after careful washing and rinsing as specified [35],
All other conditions were similar to the first interlaboratory precision
test.
The results of the second precision test are summarized in Table C-II,
Appendix C. The 100% PAAP medium had four flasks which had good
growth (> 25 x 106 cells/m^). One flask showed very little growth
(only 0. 19 x 106 cells/m£). It was evident that growth inhibition
occurred in this flask. This was the same flask (A4) which showed
toxicity in the first precision test. Evidently residual toxicity, which
was not removed even by the cleaning procedure, remained in this
flask and exhibited its growth inhibition during the second test. This
flask was discarded and the results for this flask were ignored in all
further comparisons.
For the 30% PAAP concentration (B series of flasks, Table C-II,
Appendix C) none of the flasks showed residual toxicity from the first
precision test. For the 10% PAAP medium (C series), however, three
flasks showed residual toxicity effects (Ci, C3, and C4). The two
other flasks had terminal cell numbers in excess of 1. 7 x 106 cells/ml!.
79
-------
The results of the flasks showing residual toxicity were ignored in all
further comparisons. The 1% PAAP medium (series D) showed
appreciably more growth than in the first precision test (20 x 103
cells/mi versus 2. 0 x 103 cells/m£). These levels were still too low
for accurate suspended solids measurements and because the nutrient
levels in 1% PAAP medium are on the borderline of the sensitivity
of batch assays, the results of the 1% PAAP medium assays were
ignored in all further comparisons.
A
Maximum Cell Concentration, X.
Relation with Concentration of Nutrients. One of the basic require-
ments of an algal assay is that there should be a proportional
increase in the maximum cell concentration, &, with an increase
in the concentration of the maximum cell concentration limiting
nutrient. Only then can a higher 5c be interpreted as indicating
a higher concentration of the maximum cell-limiting nutrient.
A A
X Based on Suspended Solids. The response of the X based on
suspended solids for the 10, 30, and 100% concentrations of
PAAP medium was significantly correlated (p = 0. 99) with the
concentration of PAAP medium in accordance with the following
linear equation which is shown in Figure 18:
X (suspended solids, mg/l) = 50.43 + 3. 31 (% PAAP), (49)
However, according to theory, the intercept for such a relation-
ship should be zero because no growth should occur in the absence
of nutrients. Nutrient carryover with the inocular might result
in the presence of a small positive intercept but certainly it should
not be as high as 50 mg/£. The intercept of the regression is far
in excess of what is expected. One reason for this anomaly might
be that something else other than nutrients in the medium was
limiting the maximum growth in the 100% PAAP medium cultures.
This would result in a depressed regression line with a large
intercept.
To test the above hypothesis, a regression equation was computed
omitting the 100% PAAP values and using only the values for the
10% and £0% PAAP medium cultures. The following relationship
between X (suspended solids) and concentration of PAAP medium
was obtained.
X (suspended solids, mg/f) = 0. 76 + 5.45 (% PAAP). (50)
80
-------
This relationship was highly significantly correlated (p = 0. 99,
n = 7). It is obvious that the inclusion of the 100% PAAP values
depressed the regression line. It appears that exclusion of
these values in Equation 50 best represents the response of
Selenastrum capricornutum to PAAP medium. It should be noted
that both Equations 49 and 50 have higher regression coefficients
than those for the relationship reported by Porcella et_al. [34].
That relationship also had a high positive intercept (30.^9) which
also indicates that at the higher concentrations of the PAAP
medium something other than nutrients was limiting growth.
% PAAP MEDIUM
FIGURE 18. RELATIONSHIP OF Selenastrum capricornutum
RESPONSE (SUSPENDED SOLIDS) TO
CONCENTRATION OF PAAP MEDIUM
X Based on Cell Numbers. A highly significant fp = 0. 99)
correlation (r = 0. 98) was observed between X based on cell
numbers and concentration of PAAP medium (10%, 30%, and
81
-------
100%). The regression equation which expresses this relation-
ship is shown on Figure 19 and is as follows:
& (number of cells, 106/m0 = 0. 319 (% PAAP) - 1. 13. (51)
If the 100% PAAP cultures are omitted the regression becomes:
X (number of cells, 106/m{) = 0. 314 (% PAAP) - 1. 0. (52)
These two equations are virtually identical and based upon cell
members the response to the 100% PAAP cultures were proportional
to those of the 10% and 30% cultures. Both regression equations
had significantly higher regression coefficients (0. 319 and 0. 314
versus 0. 092) than those for the relationship reported by Porcella
et aL [34], The difference cannot be explained except that two
different strains of Selenastrum capricornutum were used and it
may reflect some of the unexplained anomalies associated with
the bottle assay.
30 40 50 60 70
% PAAP MEDIUM
100
FIGURE 19. RELATIONSHIP OF Selenastrum capricornutum
RESPONSE (CELL NUMBERS) TO
CONCENTRATION OF PAAP MEDIUM
82
-------
With X based on cell numbers there was no depression of the
^egression line with a resulting increase^in the intercept as for
X based on suspended solids. From the X based on cell numbers
it would appear that nothing but the nutrients in the culture media
limited the maximum growth of the alga. This is in direct contrast
to the results of & based on suspended solids, which again reflects
the variabilities associated with bottle assays.
A
X Based on Total Cell Volume. A highly significant (p = 0. 99)
correlation coefficient (r = 0. 97, n = 11) was observed between
X based on total cell volume and concentration of PAAP medium
(10%, 30%, and 100% PAAP). The regression equation which
expresses this relationship is shown on Figure 20 and is as
follows:
X (total cell volume, |x3/mO = 19. 34 (% PAAP) - 24. 2 . (53)
Again if the 100% PAAP culture values are not taken into account
the relationship is:
X (total cell volume, |i3/m£) = 19. 93 (% PAAP) - 38. 0 . (54)
These two equations are very similar and the response of the
100% cultures were nearly proportional to those of the 10% and
30% cultures. The inclusion of the 100% culture values did not
result in a depression of the regression line with a resultant
increase^in the intercept. In this regard these values agree with
those of X based on cell numbers and not with those of it based on
suspended solids. However, a negative intercept of 0% PAAP
(-24 and -38) is difficult to explain on a theoretical basis.
A A
Precision of X Estimates. The precision of the analyses for X
for the 100% and 30% PAAP cultures was estimated and is
summarized in Table IX.
The suspended solids measurements had the lowest coefficient of
variation (approximately 5%) while the cell numbers and total cell
volume estimates had coefficients of variation of approximately
13% and 11%, respectively. These values are in the same range
(approximately 10% to 15%) as the values reported by Porcella
et al. [34], Therefore, the coefficient of variation of batch algal
asYays appears to be in the order of 15%: a precision which seems
to be acceptable in a biological assay technique.
83
-------
2000
E
N
JO
- 1500
UJ
s
3
_i
O
>
_J
UJ
o
1000
{2
o
-------
TABLE IX
MEAN VALUES AND PRECISION ESTIMATES FOR X
BASED ON DIFFERENT BIOMASS PARAMETERS
Maximum Cell
Concentration
A
(X) Based on
100% PAAP
30% PAAP
10% PAAPb
Mean
Mean
cv(%)a
Mean
cv(%)a
Suspended solids,
mg / i
377
5. 5
164. 0
4. 8
55. 3
Cell numbers,
106/mtf
30, 8
12. 6
8. 4
13. 2
2. 1
Total cell
volume,
106 |i3/ml
1910, 0
12, 7
560, 0
10, 1
161. 0
aC (%) = coefficient of variation = 100 (standard deviation)/mean.
^Coefficient of variation could not be determined because of
insufficient number of replicates.
Discussion and Summary of Maximum Cell Concentration (X)
Response, The precision of the X response based on any of the
biomass parameters, i. e. , suspended solids, cell numbers, or
total cell volumes, was acceptable (Cv ^ 15%) for a biological
assay technique. However, there appeared to be a difference in
the type of response from the different biomass parameters, i, e. ,
a growth limitation by something other than the culture nutriments
was observed with & based on suspended solids but not with X
based on either cell numbers or total cell volumes.
The difference in behavior can be explained if the mass per cell
decreased with increased nutrient concentrations in the culture
media. An examination of the mean cell volumes showed that the
mean cell volume of Selenastrum capricornutum in 100% PAAP
was 62 (j,3/cell (C = 3. 3%); this increased to 66. 9 (J.3/cell (Cv =
7. 1%) for 30% PAAP and to 79. 5 m-3/ cell (Cy = 18. 3%) for 10%
PAAP, respectively. The regression line, i. e. ,
Mean cell volume (|jl3/cell) = 74. 3 - 0. 13 (% PAAP) (55)
which describes this relationship had a significant correlation
coefficient (r = 0. 55, n = 22, p = 0. 99K If the density of cells
85
-------
grown in different concentrations of PAAP medium is constant,
then the mass/cell decreases as the nutrient concentration in
the culture medium increases. Therefore a difference in the
regression between & based on suspended solids and on cell
numbers should be observed. Variations in the cell size of
Selenastrum capricornutum under different culturing conditions
(with or without carbon dioxide addition) have been reported also
by Dixon and Scherfig [55],
A
The significant relationship between X based on total cell volumes
and concentration of PAAP medium indicates that the effect of
cell size variation was not sufficient to account for the difference
in X response. If it were sufficient then the ba^sed on total cell
volumes should have shown the same relation as X based on
suspended solids. However, the higher precision of the suspended
solids measurements suggests that the depressed values for 100%
PAAP were not artifacts but were due to a growth limitation by-
something other than the nutrients in the culture media. It is
possible that either carbon dioxide or light may have limited the
growth in the higher nutrient concentration samples. Because
it has been shown that the use of 800 ft-c light intensities showed
no significant differences from the use of 400 ft-c intensities
[58] it is doubtful if light limitation was experienced. Therefore,
carbon limitation was probably responsible for the anomaly.
Carbon dioxide is a part of the carbon dioxide-bicarbonate-
carbonate buffer system of waters, i. e. ,
COz (air) OH" OH"
u + +
HzO + COz (aq) ^H2C03ri H+ + HCO"3 ^ C03" + H+ (56)
U 11 11
COz (algae) HzO HzO
In the case where algae are utilizing C02 at a higher rate than at
which it can be replenished from the air, bicarbonates and
carbonates will be converted to COz (aq) resulting in the production
of hydroxyl ions and a concomitant rise in pH. Examination of the
pH values of the cultures (Table C-II, Appendix C) shows that the
pH in the 100% cultures (A series) was 8. 6 or above on the 9th
day. The pH in the culture which had little growth because of
toxicity (A4) reached only 8.4 at maximum. The 10% and 30%
PAAP cultures reached maximum pH values of 8, 0 and 8, 5,
respectively.. Therefore, the greater the growth in the cultures,
the higher the maximum pH observed,. This also suggests that
carbon limitation could have occurred in the 100% cultures.
86
-------
Carbon limitation during the assaying of high nutrient concen-
tration samples is a distinct possibility. If carbon is not the
nutrient of interest, extreme care must be exercized to prevent
carbon limitation. The maintenance of culture pH values between
7. 0 and 8. 3 through the use of correct surface/volume ratios in
flasks or through the use of ventilation systems has been recom-
mended [77]. However, the above results indicate that even with
a ventilation system in operation, pH values in high-nutrient
concentration samples can increase sufficiently to experience
carbon limitation. Increased air flow rates are necessary to
prevent such limitations, and it appears that the use of flasks
with specified surface/volume ratios is no guarantee that carbon
limitation will not occur in high nutrient concentration samples.
In such samples, e, g. , undiluted secondary effluents, the pH of
the cultures must be monitored frequently and be kept in the
range 7. 0 to 8. 3,
One point of caution should be noted in interpreting the foregoing
data as to the adequacy of the batch assay using the maximum cell
concentration, X, as the response parameter with respect to
nutrient concentrations (% PAAP medium). The lowest concen-
tration of nutrients used (10% PAAP) resulted in a maximum cell
concentration, 5c, of about 55 mg/( of cells at the end of 21 days
incubation. This concentration of cells is much higher than
normally encountered in highly eutrophied waters with the possible
exception of a severe bloom condition. As this is near the lowest
level of sensitivity of the batch assay and all other nutrient concen-
trations assayed were higher than 10% PAAP ~ 62 fig P/l, it is
questionable that these correlations are very indicative of the
accuracy of the batch assay for assaying natural waters with
moderately low concentrations of nutrients (i. e. , « 60 fig PfI,
« 140 fag N/0- As was noted earlier, the results of the batch
assays on the 1% PAAP medium (6 p.g P/0 and 14 |j.g N/f) were
omitted from the regression analyses because of the lack of
sensitivity at this level of nutrients. It may be that the most
important range of nutrient concentration that need to be assayed
are in the range equivalent to 1% to 10% PAAP medium — and this
is the range of concentration where the batch assay has yet to be
shown to be interpretative or applicable.
Maximum Specific Growth Rates, [jl^. The mean maximum specific
growth rate (based on cell counts) of Selenastrum capricornutum
in the 100% PAAP medium was 1. 95 day"1 with a coefficient of
variation of 14. 6%. For the 30% PAAP it was 2. 06 day-1 with a
coefficient of variation of 15. 8%. The value for for the 10%
PAAP was 1. 53 day-1; and due to the lack of sufficient number of
replicates an estimate of the variance could not be made. However,
the results seem to indicate that there was no difference in the
growth rates of the 100% and 30% PAAP media and that a somewhat
87
-------
lower growth rate was obtained in the 10% samples. Total cell
volume estimates could not be used to determine because the
instrument for determining mean cell volumes was under repair
at that time.
The absence of large or significant differences between growth
rates (p^) obtained in different strength samples should not be
interpreted as growth rate being an inferior response measure-
ment. Differences in the growth rates for an organism is dependent
on the ability of the organism to utilize a rate-limiting nutrient
efficiently at very low concentrations. This ability is reflected
by the half saturation constant, K , of the organism for a specific
nutrient, i, e. , the concentration of the nutrient at which the
organism will attain half the maximum specific growth rate (pi).
In the case of PAAP medium it appears likely that phosphorus is
the primary limiting nutrient. Porcella e^al^ [34] reported that
the Ks value for phosphorus for Selenastrum capricornutum was
less than 25 (a.g P11 while later in this report evidence is presented
that the Ks for phosphorus is as low as 3-6. 0 (jig P/L Therefore,
in PAAP medium, with phosphorus being the limiting nutrient,
differences in growth rates for Selenastrum capricornutum cannot
be observed unless the phosphorus concentration is reduced to
very low levels, i. e„ , less than about 2 to 3 times the Ks value
or about 6 to 15 fag/2 as P„ This is markedly less than the P
concentration in 10% PAAP (62 fig/0) so that no differences in
growth rates would be expected except for other limiting factors
between the 10%, 30%, and 100% PAAP concentrations.
Natural Water Assays
Batch algal assays of Clear Lake and Sacramento River water samples
were undertaken to determine: 1) the linearity of response and the
precision of maximum cell concentration (X) in the water samples,
2) the effect of dilution on the growth rate (ja^), 3) the effect and
linearity of response of the addition of secondary effluent addition to
samples, and 4) the identification of the growth-limiting nutrient in
the samples.
Water samples were obtained and were pretreated either by pasteuriza-
tion or autoclaving. Thereafter the water sample^ were divided into
the necessary aliquots. Linearity of response of X and also the
presence of toxic materials were investigated by dilution of Clear Lake
or Sacramento River water with distilled deionized water to concen-
trations of 33, 3% and 66, 7% (V/V). These dilutions as well as 100%
Clear Lake water were assayed in triplicate. The effect of dilution
on the algal growth rate was also investigated in these dilutions.
Addition of secondary effluent in concentrations of 10%, 20%, and 30%
(V/V) to undiluted Clear Lake or Sacramento River water was used
to determine the effect of addition of secondary effluent. Triplicate
88
-------
assays were made on each cconcentration of secondary effluent, and
the effect of secondary effluent addition on the algal growth rate was
also followed. Enrichment (spiking) procedures, aimed primarily at
nitrogen, phosphorus, and iron were used to determine the growth-
limiting nutrient. Micronutrients and macronutrients were respectively
used as combined spikes. The spiking procedure is summarized in
Table X.
TABLE X
EXPERIMENTAL PROCEDURE FOR IDENTIFICATION
OF GROWTH-LIMITING NUTRIENT
Enrichment (Spike) of
Clear Lake Water
Replicate Flasks
None
3
Na . . . .
2
b
p
...... 2
FeC
...... 2
N, P . .
2
N, Fe
...... 2
P, Fe
2
N, P, Fe . . . .
2
N, P, Fe, Micronutrients , . .
2
30% PAAP medium0 ........
2
Secondary effluent"^ . ,
3
Control®
1
cL • •
Spiked with NaN03, final concentration after spike
2. 5 mg N/£ + [N] of sample.
¦L
Spiked with KHzP04, final concentration after spike
200 [Jig P/0 + [P] of sample
CSpiked with FeCl3, final concentration after spike
20 fig Fe/f + [Fe] of sample,
Micronutrient spike identical in composition and
concentration to that of 30% PAAP medium.
Final concentration of nutrients after spike equal to
30% PAAP medium,
"^Secondary effluent from pilot plant using activated
sludge procedures.
^Control was 30% PAAP medium.
89
-------
All spike concentrations were selected to increase X with at least 50
mg/f (suspended solids) based upon the yield coefficients presented by
Porcella et a_L [ "34J „
Clear Lake Studies. During the Clear Lake assays control flasks of
_ . ... i i i '—A y
30% PAAP medium produced a X ol suspended solids of 1.48. 3 mg/!
(Cv = 4. 6%) and of cells x lO^/mf ol 3. 164 (Cv - 9.. 2%), In addition,
in Clear Lake water supplemented with 9% PAAP medium an X of
suspended solids of 102. 5 mgfl (Cv ~ 8. 5%) and of cells x 106/m? of
2, 24 (C - 4. 9%) were produced. Therefore, the inoculum appeared
to be responsive to nutrients and growth occurred as expected.
A
Maximum Cell Concentration (X) and Toxicity. Maximum cell
concentration (5Tj based on suspended solids, cell counts, and
total cell volume were determined. The growth curves for all
assays are presented in Table C-lll, Appendix C.
Undiluted Clear Lake water produced 39. 1 mg/f suspended solids
(Cv ¦= 6. 3%), Clear Lake water diluted to 66. 7% produced 27. 4
mg/? suspended solids, but the Clear Lake water diluted to 33. 3%
produced nearly the same (26, 6 mg/f). This higher-than-expected
value in the 33, 3% samples cannot be explained. The best fit
relationship between suspended solids and the 66, 7% and 100%
cell concentrations is shown in Figure 21,
Undiluted Clear Lake water produced 0. 573 x 106 Selenastrum
cells/m^ (Cv •- 6. 7%). Correspondingly, the 66. 7% dilution
produced 0.452 x 1 06 eel is/mf. There was a highly significant
relationship (r = 0„ 97, n - 9) between the dilution of Clear Lake
water and the X (cell counts) as shown in Figure 22. Similarly,
a highly significant (p ^ 0, 99) relation was observed between X
based on total cell volume and the dilution of Clear Lake water
(r = 0. 9, n - 9) as shown in Figure 23.
Dilution of Clear^ Lake water appeared to yield a. proportional
decrease in the X based on suspended solids, cell counts, or total
cell volume (Figures 21, 22, and 23). No evidence was obtained
of the presence of toxicants as observed by Porcella et al, [34],
The toxicity observed in that study must have developed upon the
prolonged storing of samples, even in the dark at 2°—4°C. The
prolonged storage of water samples before assaying should be
avoided.
No ready explanation is available for the lack of linearity of
response between X and percent sample when & was based upon
the cell concentration parameter, suspended solids. Neglecting
the values for the 33% samples, the ordinate corresponding to 0%
was still about 15% of the response observed for the 66% sample.
90
-------
CLEAR LAKE WATER (%)
FIGURE 21. RELATIONSHIP BETWEEN DILUTION OF
CLEAR LAKE WATER AND X BASED ON
SUSPENDED SOLIDS
33.3 65 7
CLEAR LAKE WATER (%)
100
FIGURE 2Z. RELATIONSHIP BETWEEN DILUTION OF
CLEAR LAKE WATER AND X BASED ON
CELL NUMBERS
-------
CLEAR LAKE WATER (%)
FIGURE 23. RELATIONSHIP BETWEEN DILUTION OF
CLEAR LAKE WATER AND & BASED ON
TOTAL CELL VOLUME
Yet, the growth response was linear with concentration for X
based upon cell numbers and cell volume. However, in both
cases the 0% sample growth response (ordinate) varied from
about 15% to as high as 35% of the response for the 100% sample.
No explanation for this anomaly is available.
Precision of Biomass Estimates. The mean values and the
precision of the biomass estimates are summarized in Table XI.
The coefficient of variation of the different maximum cell concen-
tration estimates were all less than 11%. The precision of all
biomass measurements were adequate although only 3 replicates
were used for each of the dilutions. Consequently, the use of 3
replicates per sample during assays is recommended.
92
-------
TABLE XI
MEANS AND PRECISION OF MAXIMUM CELL CONCENTRATION
ESTIMATES IN CLEAR LAKE WATER
Maximum Cell
Production
Based On;
Clear Lake Water
33., 3%a
66. 7%
100%
Mean
c k
V
Mean
Cv
Mean
Cv
Suspended solids,
mg/t
26, 6
c
27. 4
11. od
39. 1
6. 3
Cell numbers,
106/m I
0. 324
8.4
0. 452
4. 8
0. 573
6. 7
Total cell
volume,
106|j.3/mf
26. 0
7. 9
37. 2
5. 1
50. 9
8.4
Diluted with distilled deionized water.
^Coefficient of variation = standard deviation (100)/mean,
cLost sample during assay prevented Cv calculation.
^No explanation is available for the Cv of X (suspended solids)
being higher than those for X (cell numbers) or (total cell volume).
Maximum Specific Growth Rate, Batch ((1^) in Clear Lake Water.
The effect of dilution on the growth rate is shown in Figure 24 and
Table XII. Dilution of Clear Lake water reduced the algal growth
rate as measured by either cell count or total cell volume measure-
ments. However, the (1^ estimates based on total cell volume
measurements were always less than the estimates based on
cell counts. The reason for this discrepancy is not apparent.
Dilution of Clear Lake water did not reduce the growth rate
linearly, e. g. , one-third dilution reduced the growth rate only
8. 3% based on cell count and 21. 1% based on total cell volume.
Two-thirds dilution reduced the algal growth rate only 24. 2%
based on cell counts and 31. 1% based on total cell volume. This
would be expected from a situation where the nutrient concentration
is in the zero order part of a first order—zero order function
between growth rate and limiting-nutrient concentration (e. g, ,
M-M, M equation). Dilution would not be expected to reduce the
growth rate linearly. Clear Lake water must be diluted appreciably
( > 66%) before the algal growth rate is appreciably (> 30%) reduced.
93
-------
»>
o
TJ
X
o
£
00
1.4
1.2
- 1.0
LlI
l-
<
q:
s
o
s
3
s
x
«3
s
0.8
0.6
0.4
0.2
/
~
/
/ /
/ /
u/ /
A BASED ON CELL COUNTS
o BASED ON TOTAL CELL
VOLUME
1
33.3 66.7
CLEAR LAKE WATER (%)
100
FIGURE 24. MAXIMUM BATCH GROWTH RATES ((L )
IN CLEAR LAKE WATER
94
-------
TABLE XII
MEANS AND PRECISION OF ALGAL GROWTH RATE
ESTIMATES FOR CLEAR LAKE WATER
Maximum Specific Growth Rate,
(o,^ day-1 Based On
Sample
Cell Counts
(106cells/mf)
Total Cell
Volume
(10V3/m.O
Mean
cv%a
Mean
cv%a
100% CLb
1. 22
9. 0
o
o^
»
o
6. 5
67% CL
1. 10
15. 0
0. 71
29. 1
33% CL
0. 91
4. 1
0. 62
6. 5
CL + 9% PAAP°
2. 24
5, 4
1. 91
3. 9
30% PAAP
3. 16
9. 0
2. 82
4. 8
CL + 10% SEd
3. 09
2.4
2. 66
4. 9
CL + 20% SE
3. 05
6. 1
2. 50
15. 3
CL + 30% SE
2. 82
4. 8
2. 38
6. 8
St
C % = Coefficient of variation = standard deviation
(100)/mean.
^CL = Clear Lake water.
CPAAP = PAAP medium as given in Tables I and II.
SE = Secondary effluent.
The coefficients of variation of the pL estimates were less than 15%
for all cases except one, which is only slightly higher than that
observed for X.
Utilization of Inorganic Nutrients During Assays. The chemical
composition of inorganic nitrogen and pnosphorus compounds in
the Clear Lake samples during the assays is summarized in
Table XIII. All values are the means of triplicate analyses
except for the nitrate and nitrite nitrogen (N03-N and NOz-N)
measurements in which the 3 aliquots were pooled before analysis.
Nitrogen was present in the undiluted (100%) Clear Lake sample
95
-------
TABLE XIII
MEANS OF INORGANIC NITROGEN AND PHOSPHORUS CONCENTRATIONS IN CLEAR LAKE WATER
Initial Concentration,
So
Terminal Concentration, Si
(14 days
ng/0
Clear Lake
Concentration
NH3
-N
no3 + NOz-N
PO4-
P
no3 + no2-n
PO4
-P
Mean
C a
V
Mean
Mean
C
V
Mean
C
V
Mean
C
V
100%
109. 9
2. 9
00
o
387
0. 8
8. 4
65. 1
6. 0
5. 0
66. 7%
59. 9
2.0
52. 2
229
0. 7
4. 2
13. 7
4. 2
12. 1
33. 3%
37. 1
3. 3
25. 5
127. 5
0. 0
2. 5
46. 1
2. 1
14. 3
2L
Coefficient of variation = 100 (standard deviation)/mean.
-------
at a concentration of approximately 190 |j.g N/{„ Ammonia nitrogen
(NH3-N) contributed 110 figIt and N03 and NOz-N contributed 80
(jigIt. Phosphate phosphorus (P04-P) was 387 fig/?. The initial
inorganic N/P ratio of 0, 49 suggested that nitrogen rather than
phosphorus would be the growth-limiting nutrient; however, some
organic nitrogen contained in particulate material in the water
samples could have become available upon hydrolysis during the
as says.
The accuracy of dilution of the Clear Lake water was illustrated
by the highly significant (p = 0. 99) correlation coefficients between
either initial NH3-N concentrations or PO4-P concentrations and
the dilution concentration of the Clear Lake water. A correlation
coefficient of 0, 97 (n = 9) was obtained for the relationship:
NH3-N (|j.gIt) = 3, 91 + 1. 09 (% Clear Lake water) , (57)
and a correlation coefficient of 0„ 99 (n = 9) was obtained for the
relationship:
PO4-P (fig/r : 3 89 (% Clear Lake water) - 11. 5 , (58)
During the assays almost 90% of the N03 and N02-N was utilized
(removed from solution) in all dilutions of Clear Lake water.
Terminal nitrogen concentrations were therefore less than
10 fxg NIt (Table XIII), More than 98% of the PO4-P was taken
up in all dilutions resulting in terminal phosphorus concentrations
of 6 (j.g PIt or less. Unfortunately the terminal NH3-N concen-
tration was not determined, but it is expected that most of the
NH3-N was also utilized because NH3-N is a ready nitrogen source
for most algae Very substantial reductions in the concentrations
of inorganic nitrogen and phosphorus compounds was experienced
during the assays.
Yield Coefficients for Nitrogen and Phosphorus. The yield coef-
ficients (mass of cells produced/mass of nutrient removed) for
N03-N (Yjyjo -n) were respectively 540, 571, and 1150 in the
100%, 66, 7%3, and 33. 3% dilutions of Clear Lake water. If it is
assumed that all the NH3-N in the samples was also consumed for
cell production, the values would be 214, 254, and 443,
respectively, for the 100%, 66, 7%, and 33. 3% cultures. These
values are approximately an order of magnitude higher than the
values reported by Porcella et al [34]. The inverse of the
yield coefficients, i. e, , mass of nutrient/mass of cells, gives
an indication of the concentration of a specific nutrient in the
cells. Inversion of the YN values for the different dilutions
suggests that the cells contained respectively only 0.45%, 0, 39%,
and 0. 23% nitrogen on a dry weight basis. Compared to the values
97
-------
of 3% to 5% reported by Porcella ^t_aL_ [34] and 1. 77 to 9. 43%
nitrogen reported [110] for various algae. The yield values of
the present study appear to be extremely low. The only explanation
for this is that organic nitrogen present in the particulate matter
in the water samples was hydrolyzed and used for cell production
during the assays. However, it should be noted that according to
Holm-Hansen et al. fill] the concentration of nitrogen in nitrogen-
starved cells can be very low.
Because of the relatively high mass of cells produced per unit of
inorganic nitrogen and because of the possibility of organic nitrogen
being hydrolyzed and made available for cell synthesis, it cannot
be established definitely whether or not nitrogen was the limiting
nutrient in the Clear Lake samples. However based upon the
available information, it appears than N was limiting.
The Yp values observed, i. e. , 103, 122, and 212, for the 100%,
66. 7%, and 33. 3% samples respectively, were lower than those
of Porcella et al. [34], i. e. , 800 to 1000, for phosphorus-limiting
conditions. Inversion of the Yp values suggests that the cells
contained respectively 0. 97%, 0. 82%, and 0.47% phosphorus on a
dry weight basis. These values are in the range of 0. 52% to 1. 89%
phosphorus (dry weight) for various algae, compiled by McGauhey
et al. [31], However, they are much greater than the values of
approximately 0. 1% observed [34] when phosphorus was the
growth-limiting nutrient. The data indicate that phosphorus did
not limit algal growth in the Clear Lake samples.
The use of yield coefficients to aid in identifying growth-limiting
nutrients in batch assays appears to be limited. This is due to
large variations in the yield coefficients due to excess uptake such
as observed by Porcella et al. [34] and in this study coupled with
the possible interaction of organic and inorganic forms of nutrients
such as nitrogen.
Secondary Effluent Addition to Clear Lake Water.
A A
Maximum Cell Concentration (X). The X based on three different
biomass parameters were increased by the addition of secondary
effluent to Clear Lake water. The growth curves are summarized
in Table C-IV, Appendix C.
In all instances a highly significant (p ^ 0. 99) linear relationship
was observed^between the amount of secondary effluent added and
the resulting X. These relationships are summarized in Figures
25, 26, and 27. The following relationships were obtained.
98
-------
CLEAR LAKE WATER
FIGURE 25. EFFECT OF SECONDARY EFFLUENT
ADDITION ON X BASED ON SUSPENDED
SOLIDS OF CLEAR LAKE WATER
CELLS x IOS/ml =2.22+0.252(% SECONDARY EFFLUENT)
0 10 20 30
SECONDARY EFFLUENT ADDITION (%) TO
CLEAR LAKE WATER
26. EFFECT OF SECONDARY EFFLUENT
ADDITION ON & BASED ON CELL
NUMBERS OF CLEAR LAKE WATER
-------
SECONDARY EFFLUENT ADDITION (%) TO
CLEAR LAKE WATER
FIGURE 27. EFFECT OF SECONDARY EFFLUENT
ADDITION ON £ BASED ON TOTAL CELL
VOLUME OF CLEAR LAKE WATER
X (suspended solids, mg/O = 52. 9 + 7. 1 (% secondary effluent added),
(59)
X (cells x 106/m£) = 2. 22 + 0. 252 (% secondary effluent added)
(60)
and
X (total cell volume, 106|j.3/m?) = 56. 2 + 36.4 (% secondary effluent
added).
(61)
100
-------
The above equations give an indication of the amount of secondary-
effluent that can be added to Clear Lake water before specific
limits for X are exceeded. For instance, if the limit for X is set
at 100 mg suspended solids/2, the maximum addition of secondary
effluent would be 6, 6%, However, this does not indicate what the
resultant algal cell concentration would be in Clear Lake if 6. 6% of
secondary effluent were added. It must be realized that the selec-
tion of limits for & would have to be derived empirically first
because virtually no information exists on the relationship between
the levels of biomass produced in algal assays, &, and those pro-
duced in the identical water under natural conditions. In nature the
conditions of temperature, light intensity, aeration, etc. vary
greatly and are obviously not the same as those of algal assays.
Furthermore, predation and settling phenomena tend to reduce
natural standing crops, while this is prevented in assays. There-
fore, the biomass levels in identical waters under natural and assay
conditions should invariably result in lower concentrations of cells
in the natural waters. The determination of the relationships
between 5£ and concentration of cells that would result in a specific
receiving water for given concentrations of nutrients or wastes
requires much additional research and field investigation. At
present this appears to be a gigantic assignment.
It must be stressed that the control of eutrophication problems in
a specific water system will require identification of the growth-
limiting nutrients of the receiving water and the receiving water-
waste mixture. Only then is it possible to know which nutrient, if
any, and how much must be removed from the waste stream before
discharge of wastes can occur without deleterious effects. Knowledge
of the kinetic constants of various algae for the nutrient to be
removed will be needed to decide on the amount of the nutrient that
must be removed. Therefore, the development of the relationships
between ic and the concentration of wastes for a specific receiving
water is merely a tool and not a final answer in eutrophication
analysis.
Maximum Specific Growth Rate, Batch (p^). Addition of secondary
effluent to Clear Lake water increased markedly the above that
of undiluted Clear Lake water as shown in Figure 28 and Table
XII. With a concentration of 10% secondary effluent the increase
in (1^ was approximately 2. 5 to 3. 0 times (from 1. 2 to 3. 1 day"1
for estimates based on cell counts and from 0. 90 to 2. 7 day-1 for
estimates based on total cell volume). Increased concentrations
of secondary effluent did not increase further in fact, pL^
appeared to decrease. This decrease indicates that a concen-
tration of secondary effluent above 10% is toxic to Sele^astrum
capricornutum. This toxicity was not reflected in the X of the
assays apparently because during the long incubation period the
algae had sufficient time to reach their although at a lower
101
-------
rate than with the 10% secondary effluent cultures. The use of
X alone as the algal growth parameter in assays would not detect
such toxicity; however, the toxicity is quite apparent when growth-
rate determinations (pi^) are used.
SECONDARY EFFLUENT ADDITION (%) TO CLEAR LAKE WATER
FIGURE 28. EFFECT OF SECONDARY EFFLUENT ON THE
ALGAL GROWTH RATE OF CLEAR LAKE
WATER
102
-------
In the above case there would appear to be a greater chance of an
algal bloom occurring in the presence of 10% secondary effluent
than in the presence of 30% secondary effluent, assuming the
toxicity is conservative. This is because the algal standing crop
in a water is the sum total of the different rate reactions as shown
in Figure 1, some of which increase and some which decrease the
standing crop; therefore, a reduction in the growth rate, even
through toxicity, likely would reduce the chances of an algal bloom
occurring. However, induced toxicity is not a remedy for the
eutrophication problem because as soon as the toxicant is suitably
diluted or removed the growth rate would increase and a bloom
would likely occur.
The detection of toxicity by the measurement indicates the value
of this algal response parameter. The measurement of [jl^ in batch
assays must always be included.
Higher values again were found for estimates based on cell
counts as compared to estimates based on total cell volume. The
reason for this difference is unknown. However, the behavior pat-
tern of both estimates were similar.
Enrichment (Spiking) Studies on Clear Lake.
The spiking procedure outlined in Table X was followed. The
cultures were inoculated and incubated under standard PAAP
conditions for 16 days and the & estimates were made. Daily cell
counts between the 2nd and 8th day were used for the calculation
of ph' The results of the enrichments are summarized in Table
C-V, Appendix C.
A A
Maximum Cell Concentration (X). The X estimates for the three
biomass parameters are presented in Figure 29. None of the
single nutrient spikes increased the by more than a few percent.
The N-P spike showed higher cell counts and total cell volume
estimates than the control (no enrichment made) but the suspended
solids value was lower. A small increase in all three biomass
parameters was observed with the N-Fe spike whereas the P-Fe
spike did not increase the St estimates above those of the control.
Combined spiking of the N, P, and Fe increased the & slightly,
whereas the addition of micronutrients to these three nutrients
increased the X approximately twofold or more.
On basis of the above results it is difficult to determine which, if
any, of N, P, gr Fewas a growth-limiting nutrient. The significant
stimulation of X by the addition of micronutrients indicated that a
micronutrient was a secondary growth-limiting nutrient if not the
primary growth-limiting nutrient.
103
-------
E
UJ
1500
3
^ 1000
UJ
<_)
f?
p
500
E
12
O
X
8
CO
_l
UJ
O
4
0
$
o>
e
300
CO
Q
_1
P
200
CO
o
llJ
Q
z
100
UJ
Q.
to
3
CO
0
nfnnnl inn
D.
23456789 10
1
CLEAR LAKE WATER 7
+ P, Fe
2
+ NITROGEN (N) I2.5mg/!f)8
+ N,P,Fe
3
+ PHOSPHORUS (P)(200/«g/4) 9
+ N, P, Fe, MICR0NUTRIENT5
4
+ IRON ( Fe)(20/tg/i) 10
t 30% PAAP
5
+ N, P II
30% PAAP
6
+ N, Fe
FIGURE 29. EFFECT OF ENRICHMENT (SPIKING) ON
& OF CLEAR LAKE WATER
104
-------
Maximum Specific Growth Rate, Batch The values
obtained in the spiking tests are summarized in Table XIV. The
three single nutrient spikes increased the (2-^ whereas the N-P
spike did not increase the (2^. However, no clear pattern of
behavior was observed because combined spikes such as N-P or
N-P-Fe showed lower pL values than for single nutrient spikes
themselves. Addition of micronutrients showed a higher than
for single or combined spikes of N, P, or Fe, The pattern
observed from the determinations were similar to those for &
estimates, and it is not possible to determine which nutrient was
the primary limiting nutrient. However, the stimulation by micro-
nutrients indicated that one of the micronutrients was likely
limiting growth. If a micronutrient limited growth, the degree to
which the chemicals used for spiking were contaminated by this
micronutrient is important. Because all chemicals are usually
contaminated bv traces of other chemicals it is possible that the
stimulation of X or by other nutrient spikes could have been
due to the presence of traces of a micronutrient in the nutrient
spikes.
The results of the enrichment studies of Clear Lake water did
indicate clearly which nutrient was primarily limiting growth.
To determine this further studies would have to be undertaken.
However, it appeared that one, or more, of the micronutrients
was limiting growth. The possible role of micronutrients should
not be overlooked as one of the important factors in controlling
eutrophication of natural waters.
Sacramento River Studies. During the Sacramento River assays,
control flasks of 30% PAAP medium produced an (suspended solids)
of 151. 2 mg/( (Cv = 7.4%) and an St (cells x 106/mf) of 5. 25 (Cv =
27. 0%). In addition, in Sacramento River water supplemented with
9% PAAP medium, an ft (suspended solids) of 68. 2 mg/,f (Cv = 5. 4%)
and an X (cells 106/m£) of 4.43 (Cv = 13. 3%) were produced. Therefore,
the inoculum appeared to be responsive to nutrients and growth
occurred.
Maximum Cell Concentration (X). Maximum cell concentration
based on suspended solids, cell counts, and total cell volume were
determined. The growth curves for all assays are presented in
Table C-YI, Appendix C.
A
Undiluted Sacramento River water produced an X (suspended solids)
of 15. 2 mg/i (Cv = 13. 5%). Dilution of the water to 66. 7% produced
an X (suspended solids) of 10.4 mg JI (Cv = 3. 8%), and dilution to
33. 3% produced an X (suspended solids) of 5. 8 mg/l (Cv = 2. 9%).
A highly significant (p = 0. 99) correlation coefficient (r = 0. 97,
n = 9) was observed between the 5c (suspended solids) and the
dilution of Sacramento River water as shown in Figure 30.
105
-------
TABLE XIV
EFFECT OF NUTRIENT ENRICHMENT (SPIKES) ON MEAN
SPECIFIC GROWTH RATE, BATCH (£b)
OF CLEAR LAKE WATER
Addition To
Clear Lake Water
Maximum Specific Growth Rate,
Batch (pi^)
Mean
Range
None
0. 44
0. 34 - 0. 56
Na
0. 69
0. 53 - 0. 85
Pb
0. 64
0. 62 - 0. 67
FeC
0. 70
0. 64 - 0. 77
N-P
0. 46
0. 44 - 0. 48
N-Fe
0. 79
0. 56 - 1. 03
P-Fe
0. 74
0. 64 - 0. 84
N-P-Fe
0. 54
0. 39 - 0. 69
N-P-Fe micronutrients^
0. 83
00
o
1
00
•
o
30% PAAP
1. 07
1. 00 - 1. 15
10% secondary effluent
0. 51
0. 46 - 0. 60
Control (30% PAAP)
1. 21
2L
For all N spikes 2. 5 mg NJt were added to samples.
^For all P spikes 200 fig P/i were added to samples.
For all Fe spikes 20 [ig Fe/? were added to samples.
Micronutrient spike identical in composition and concentration
to that of 30% PAAP medium.
Undiluted Sacramento River water produced an X (cells x 10°/mf)
of 0. 334 (C = 14. 5%). For 66. 7% and 33. 3% dilutions, 5c was
0. 259 x 106 cells/m? (Cy = 16. 8%) and 0. 134 x 106 cells/mf,
respectively. There was a highly significant (p = 0. 99) correlation
coefficient (r = 0. 92, n = 9) between the & (cells x 106/m{) and
dilution of Sacramento River water (Figure 31). Similarly, &
(total cell volume) values of 27. 1, 20. 0, and 9. 1 (J.3 x 106/m? for
respectively 100%, 66. 7%, and 33. 3% concentrations of Sacramento
River water were highly significantly (p = 0. 99) correlated (r =
0, 92, n = 9) with the dilution of the water as shown in Figure 32.
106
-------
20 —
15 —
lO —
5 -
t'
SSlnrig//) =1.09 + 0.141(% SACRAMENTO RIVER
WATER)
I
20 40 60 60 100
SACRAMENTO RIVER WATER (%)
FIGURE 30. RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND 5<
BASED ON SUSPENDED SOLIDS
SACRAMENTO RIVER WATER (%)
FIGURE 31. RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND
BASED ON CELL NUMBERS
-------
FIGURE 32. RELATIONSHIP BETWEEN DILUTION OF
SACRAMENTO RIVER WATER AND St
BASED ON TOTAL CELL VOLUME
As for Clear Lake samples, dilution of Sacramento River samples
yielded X values proportional to the dilution factor. No evidence
was obtained indicating the presence of toxicants. It appears that
linear X responses can be obtained for dilution of natural samples
having maximum values of ft as low as 15 mg/t suspended solids
or 0. 334 x 106 cells/m? or 27. 1 fx3 x 106/mi total cell volume.
Based upon these data it appears that the batch assay can then be
used for assaying waters with nutrient concentrations as low as
(Nfy-N + N03-N = 133 (Jig/? and PO4-P = 34 \ig/i). The linearity
in X responses indicates that the b^tch assay responds propor-
tionally to the concentration of the X limiting nutrient and that
the method can be used for screening natural samples. However,
it should be noted^that for zero concentrations of Sacramento
River water, the X response ranged from 6% to 15% of the
response for 100% Sacramento River water. No explanation for
this anomaly is available.
108
-------
The coefficients of variation for the different X estimates were all
less than 18. 5% which is acceptable. The same analyses for the
Clear Lake estimates were 11% or below. The higher value for
the Sacramento River assays is probably due to the fact that this
water contained less nutrients and therefore resulted in lower ft
estimates. The precision of these estimates for Sacramento
River water are judged to be adequate.
Maximum Specific Growth Rate, Batch (£b)- The effect of dilution
on the of Sacramento River assays is shown in Figure 33.
Dilution to 66. 7% and 33. 3% reduced the jjL^ (cell counts) by 37%
and 79%. respectively, and the jl^ (total cell volume) by 17% and
66%, respectively. Dilution of Sacramento River water had an
appreciable effect on the of the samples, probably because
Sacramento River has a moderately low nutrient content, at
least as compared to Clear Lake water. Hence dilution of
Sacramento River water has a greater effect on the than dilution
of Clear Lake water because it brings the nutrient content of the
diluted samples down into the range where growth is first order
with respect to nutrient concentration.
The precision of the Sacramento River estimates is presented
in Table XV. Inspection of Table XV indicates that the coefficients
of variation (Cv) |or the estimates of were markedly higher than
for estimates of X. In fact, the middle dilution of Sacramento
River water (67%) had Cv's of 43% and 57% for the ji-^ based upon
cell count and cell volume, respectively. Most of the other values
of the Cv were less than 20%. No adequate explanation is available
for this lower level of precision except for the inherent variability
in the growth rate of the batch assay and the fact that the counts
used for computation of were much less than for ft.
Utilization of Inorganic Nutrients During Assays. The concentration
of inorganic nitrogen and phosphorus compounds in the Sacramento
River samples used for these assays is summarized in Table XVI.
All reported values are the means of either duplicate or triplicate
analyses. Nitrogen was present in the undiluted Sacramento River
water at a concentration of about 133 |xg N/2; of this N03 and N02-N
contributed 81% and NH3-N 19%. Phosphate phosphorus was present
at a concentration of 34 |jLg P/t. Sacramento River contained 60
fxg N/< and 362 fig P/< less than the Clear Lake samples (Table
XIII), which explains the lower ft for the former. Sacramento
River has a much lower nutrient content than Clear Lake.
109
-------
4 -
1.2 —
a Based on cell counts
o Based on total cell volume
Total cell volume
33.3 66.7
SACRAMENTO RIVER WATER (%)
100
FIGURE 33.
MAXIMUM BATCH GROWTH RATE
IN SACRAMENTO RIVER WATER
110
-------
TABLE XV
ALGAL GROWTH RATE ESTIMATES, ft,
FOR SACRAMENTO RIVER b
Sample
Maximum Specific Growth Rate,
(i^ day-1, Based On
Ci'll Count
(106 cells/m^)
Total Cell
Volume
(106 |i3/m?)
Mean
cv%a
Mean
cv%a
100% SRb
1. 21
13. 8
o
00
-j
23. 9
67% SR
0. 77
42. 7
0. 72
56, 7
33% SR
0. 25
15, 8
0. 30
42. 5
SR + 30% PAAPC
3, 17
20. 3
2. 33
14. 0
30% PAAP
2, 25
13. 5
2. 09
16. 6
SR + 10% SEd
2. 32
17. 1
1. 67
8. 9
SR + 20% SE
2„ 50
3. 7
1. 92
4. 5
SR + 30% SE
2. 79
5. 7
2. 25
5. 3
a
C % = Coefficient of variation = 100(standard
V i
deviationjmean"1.
1^
SR = Sacramento River,
Q
PAAP = Medium given in Tables I and II.
SE = Secondary effluents.
111
-------
TABLE XVI
INORGANIC NITROGEN AND PHOSPHORUS CONCENTRATIONS IN SACRAMENTO RIVER WATER
Initial Concentration, S
Terminal Concentration, S.
*-o
t = 14 days
Sacramento River
(H-g/-)
(H-g/0
Concentration
NH3-N
no3 + NOz-N
po4
-P
nh3
-N
no3 +
no2-n
po4
-P
Mean
Mean
Mean
C a
Mean
C
Mean
C
Mean
C
V
V
V
V
100%
25. 2
107 6
34. 2
3. 2
11. 9
10. 2
12. 1
59. 0
7. 8
3. 8
66. 7%
16. 6
75. 0
22. 1
7 1
12. 6
b
5. 7
14. 3
5. 2
18. 5
33. 3%
14. 3
36. 7
9. 7
4. 7
11. 9
26. 9
8. 5
14. 3
6. 0
25. 0
aCoefficient of variation = 100 (standard deviation)/mean
^Lack of replicates prevented estimate of coefficient of variation.
-------
The use of filtration pretreatment in earlier studies of these two
waters apparently resulted in a large removal of nutrients because
the chemical analyses for both waters were similar [34], The
current findings indicate that filtration pretreatment must have
removed substantial amounts of nutrients from the Clear Lake
samples because these waters are considerably different in
chemical composition.. The initial inorganic N/P ratio of 3. 9
suggests that nitrogen rather than phosphorus would be the limiting
nutrient (if it is one of these) because phosphorus appears to be
limiting at N/P ratios of more than 30 [34].
The accuracy of dilution of the Sacramento River water was
illustrated by the highly significant (p 5 0. 99) correlation coef-
ficients between either NH3-N or PO4-P or N03 and N02-N con-
centrations and dilution concentration of the Sacramento River
water. The correlation coefficients were 0. 99, 0. 99. and 0. 86,
respectively, for the relationships:
N03 and NOz-N (|o.g/{) = 3. 3 + 1. 05 (% of Sacramento River water)
(62)
PO4-P (fxg/0 = 0. 37 (% Sacramento River water) - 2. 5 (63)
NH3-N (fig/?) = 8. 8 + 0. 15 (% Sacramento River water) , (64)
During the assays at least 79% of the NOa and NOz-N was removed
from solution in all dilutions of Sacramento River water. Terminal
concentrations were 12, 1 (jig N/? or less. In the 33. 3% samples
apparently only 17% of the NH3-N was removed from solution;
however, the amount of hydrolysis of nitrogenous particulate
matter in the samples is not known, nor its effect on NH3-N
concentrations. Because the samples contained a considerable
amount of particulate matter, release of NH3 through hydrolysis
was a distinct possibility. Removal of PO4-P was at least 38%
(33, 3% dilution) and as high as 77% (66. 7% and 100% cultures).
However, the terminal PO4-P concentrations were about the same
for all dilutions at a level of 5-8 |ig P/f. The amount of inorganic
nutrient removal was not as high as that in the Clear Lake samples;
the reason for this is not apparent.
Yield Coefficients for Nitrogen and Phosphorus. The Yj^ values
for the 100%, 66„ 7%, and 33. 3% concentrations of Sacramento
River water were respectively 140, 142, and 191. As with the
Clear Lake assays, these values were much higher than expected,
Porcella et al. [34] obtained Yj^ values of 20 to 30 at low initial
N/P ratios when nitrogen was the likely limiting nutrient. The
values found in this study are at least four times greater than
113
-------
those reported in the previous study. Inversion of the values
suggests that the cells contained only 0. 71%, 0. 70%, and 0. 52%
nitrogen on a dry weight basis. Because these values seem low,
it appears that hydrolysis of nitrogenous compounds in the
particulate matter may have supplied nitrogen for cell synthesis.
Based upon the data obtained on yield coefficients, it is not
possible to conclude whether or not nitrogen is the growth-limiting
nutrient.
The Yp values of 577, 615, and 1582 for respectively the 100%,
66. 7%, and 33. 3% concentrations suggest that phosphorus might
have been the growth-limiting nutrient because these values
approached the Yp values of approximately 900 to 1000 (when
phosphorus was growth limiting) [34], However, the fairly high
terminal PO4-P values (5 to 8 fig P/i) and the fairly low initial
PO4-P values suggest that if phosphorus was limiting growth, lower
terminal values would have been expected.
As with the assay of Clear Lake water, determination of yield
coefficients did not aid significantly in identification of the growth-
limiting nutrients. The reason appears to be the tremendous
variation in yield coefficients with batch assays. This finding
also raises the significant question as to how satisfactory the
batch assay can become in identifying the growth-limting nutrient
until a rational explanation is available for the variation in the
yield coefficient.
Secondary Effluent Addition to Sacramento River Water.
A A
Maximum Cell Concentration (X). The X based on three different
biomass parameters were increased by the addition of secondary
effluent to Sacramento River water. The growth curves are
summarized in Table C-VII, Appendix C.
In all instances a significant (p ^ 0. 99) linear relationship was
observed between the amount of secondary effluent added and the
resulting X. However, regression for & based upon cell counts
and cell volume (Figures 34 and 3 5) indicate a lot of scatter in
the data. No explanation is available for the variability in the
data. These relationships are summarized in Figures 34, 35,
and 36. The regression equations obtained were as follows:
A
X (suspended solids, mg/f) = 21. 7 + 6. 64 (% secondary effluent)
(65)
X (cells 106/m?) = 2. 53 + 0. 342 (% secondary effluent added)
(66)
& (total cell volume, 106^3/m?) = 202. 2 + 27. 14 (% secondary
effluent added) .
(67)
114
-------
These relationships are similar to those obtained for Clear Lake
water, i. e. , the slopes for Clear Lake were respectively 7. 1,
0. 252, and 36.4 compared to the slopes of 6. 64, 0. 342, and
27. 14 for the Sacramento River samples.
Using the foregoing relationships it is possible to estimate the
amount of secondary effluent that can be added to the Sacramento
River water before a specific limit is reached. For instance,
11. 76% secondary effluent will produce an & of 100 mg suspended
solids/?, compared to 6. 6% in the case of Clear Lake. The dif-
ference for the two waters can be attributed to the lower nutrient
content of the Sacramento River as indicated by the higher zero
intercept of the Clear Lake equation (52. 9 versus 21. 7), and the
lower slope for the Sacramento River (6. 64 versus 7. 1). However,
as indicated previously how these numbers relate to the standing
crop of plankton, in either body of water is unknown. Moreover
it is just this relationship that one needs to know to provide
adequate control measures.
Maximum Specific Growth Rate, Batch The effects of
addition of secondary effluent to Sacramento River water on
are shown in Figure 37. Addition of secondary effluent increased
the pt-^ (based on cell counts or total cell volume). Unlike the
Clear Lake sample, the addition of secondary effluent indicated
no toxic effects on the algal growth rate of the Sacramento River
water. Addition'of secondary effluentjo; Sacramento River
water will increase the possibility of an algal bloom developing
and up to a concentration of 30% secondary effluent the growth
rate appears to increase.
The data reported in Figure 37 indicate that there is something
in secondary effluent (possibly trace elements) other than its
nutrient content that is lacking in Sacramento River water. Once
some of these limiting substances are added, the growth rate
increases markedly (i. e. , 0 to 10% secondary effluent). There-
after from 10% to 30% the increase in growth rate is approximately
linear with respect to the increase in concentration of nutrients
added — at least up to a concentration of 30% secondary effluent.
It would have been informative if a series of 1% or 2% secondary
effluent assays had been included to assess this observation.
Growth rate measurements based on total cell volume measure-
ments were again lower than those based on cell counts. The
reason for this is hot apparent. The precision of the (2^ estimates
for the addition of secondary effluent (Table XV) was satisfactory.
115
-------
SECONDARY EFFLUENT ADDITION (%) TO
SACRAMENTO RIVER WATER
FIGURE 34. EFFECT OF SECONDARY EFFLUENT
ADDITION ON X BASED ON CELL
NUMBERS OF SACRAMENTO RIVER
WATER
SECONDARY EFFLUENT ADDITION {%) TO
SACRAMENTO RIVER WATER
FIGURE 35. EFFECT OF SECONDARY EFFLUENT
ADDITION ON X BASED ON TOTAL C EI L
VOLUME OF SACRAMENTO RIVER
WATER
-------
SECONDARY EFFLUENT ADDITION (%) TO
SACRAMENTO RIVER WATER
FIGURE 36. EFFECT OF SECONDARY EFFLUENT
ADDITION ON & BASED ON SUSPENDED
SOLIDS OF SACRAMENTO RIVER WATER
a Based on cell counts
o Based on total cell volume
SECONDARY EFFLUENT ADDITION (%) TO
SACRAMENTO RIVER WATER
FIGURE 37. EFFECT OF SECONDARY EFFLUENT
ADDITION ON (1, OF SACRAMKVTO
RIVER WATER
-------
Enrichment (Spiking) Studies on Sacramento River Water.
The spiking procedure outlined in Table X was followed. The
cultures were inoculated and incubated under standard PAAP
conditions for 16 days and the St estimates were made. Daily
cell counts between the 2nd and 8th day were used for the
calculation of The results of the enrichments are sum-
marized in Table C-VIII, Appendix C.
A A
Maximum Cell Production (X). The X estimates for the three bio-
mass parameters are presented in Figure 38. All three biomass
parameters gave essentially similar results. Of the single^
nutrient spikes only nitrogen appears to have increased the X
above that of unspiked water. This suggests that nitrogen was the
primary limiting nutrient in the water. The increased & of the
nitrogen-phosphorus spike (whereas the nitrogen-iron and iron-
phosphorus spikes had lower X) suggests that phosphorus was the
secondary limiting ijjitrient. This hypothesis is also supported
by the high level of X of the nitrogen-phosphorus-iron spike. The
absence of any significant increase in SC between the nitrogen-iron-
phosphorus spike and the nitrogen-phosphorus spike suggests that
iron was not a limiting nutrient; however, one of the micronutrients
appears to be limiting because addition of micronutrients produced
the maximum value of St. This supports the hypothesis cited
earlier based upon examination of the data shown in Figure 37.
A
Determination of the X values for different spikes of nutrients
appears in this case to offer promise in the identification of
limiting nutrients in a specific sample.
Maximum Specific Growth Rate, Batch (p^). The values
obtained in the spiking tests are summarized in Table XVII.
The highest values were obtained in the control (30% PAAP)
and the N-P-Fe-micronutrient spikes, and were about 3. 5 times
as high as the unspiked Sacramento River water. The next
highest growth rates were respectively for the Fe, P-Fe, N-P,
N-Fe, and N-P-Fe spikes, whereas the N and P spikes had
values somewhat less than the Sacramento River water itself.
The ranges of the determinations (Table XV) showed that the
differences were not due to variations in the (1^ measurements
but to real differences that existed between the spikes as illustrated
in the summarized growth curves presented in Figure 39.
The only difference in composition between the N-P-Fe-micro-
nutrient spike and the 30% PAAP medium spike was in the presence
of macronutrients other than N and P, These macronutrients
apparently were not growth rate limiting because they were present
in sufficient quantities in the N-P-Fe-micronutrient spike. The
118
-------
E
"a
~ 800
LU
2
3
9
400
LlJ
O
_l
P
12
"b
X
tn
_i
_i
LU
O
o>
£ 300
(0
Q
Hi
O 200
cn
z
LlJ
Q.
CO
3
(f>
100
JEL
eld.
m.
JQ
m.
8
10
1 SACRAMENTO RIVER WATER
2 + NITROGEN (N) (2.5mg/£)
3 + PHOSPHORUS (P)l200pt}/£)
4 + IRON (Fe) (20/ig//)
5 + N, P
6 + N, Fe
7 + P, Fe
8 + N, P,Fe
9 + N, P, Fe, MICRONUTRIENTS
10 t 30% FfcAP
11 30% PAAP
FIGURE 38. EFFECT OF ENRICHMENT (SPIKING) ON
THE X OF SACRAMENTO RIVER WATER
119
-------
TABLE XVII
EFFECT OF NUTRIENT ENRICHMENT (SPIKES) ON
MEAN SPECIFIC GROWTH RATE, BATCH (jl)
OF SACRAMENTO RIVER WATER
Addition To
Sacramento River Water
Maximum Specific
Growth Rate,
Batch (p^)
Mean
Range
None
0. 61
0. 44 - 0. 83
Na
0. 29
0. 25 - 0. 34
Pb
0. 41
0. 37 - 0. 45
Fec
1. 95
1. 76 - 2. 15
N-P
1. 57
1. 55 - 1. 59
N-Fe
1. 50
1.49 - 1. 51
P-Fe
1. 75
1. 13 - 2. 38
N-P-Fe
1. 38
1. 42 - 1. 33
N-P-Fe-xnicronutrients^
2. 02
1. 75 - 2. 30
30% PAAP
2, 02
1. 70 - 2. 38
10% secondary effluent
0. 62
0.43 - 0. 76
Control (30% PAAP)
2. 21
aFor all N spikes 2. 5 mg N/J were added to sample.
U
For all P spikes 200 (ig 11 were added to sample.
°For all Fe spikes 20 jo.g Fe/£ were added to sample.
^Micronutrient spike identical in composition and
concentration to that of 30% PAAP medium.
120
-------
FIGURE 39. GROWTH CURVES OF ENRICHMENTS (SPIKES)
OF SACRAMENTO RIVER WATER
121
-------
differences in the p, stimulation between the N-P-Fe-micronutrient
spike and the rest ol the spikes appear to be due to the presence of
micronutrients. It is strongly indicated that the (jl^ of the Sacramento
River water was limited by micronutrients. Because the experiment
was primarily aimed at investigating and evaluating the spiking of
N, P, and Fe, micronutrients were lumped together and not
investigated singly. This is unfortunate, as it would have been
most informative to have a separate spike of micronutrients.
However, laboratory chemicals are usually contaminated with
small amounts of other chemicals; therefore, the growth stimula-
tion of spikes other than micronutrients could have been due to
the trace contaminants (micronutrients) and not to the nutrients
or other chemicals. The role of Fe might be considered in this
connection; in all cases where Fe was one of the spikes, higher
values (compared to the unspiked water) were observed. It is
not clear if Fe per se or a contaminant of micronutrient(s) in the
FeCl3 solution induced these increases,
The foregoing observations have considerable practical^implica-
tions. First, if conclusions were drawn only from the X values,
nitrogen would have been identified as the limiting nutrient of
concern. However, the fact that nitrogen did not increase (a
seeming decrease was noted) the fib of the water indicated that
even though a higher would be produced in the presence of
nitrogen, this would not occur at a higher rate. Therefore, if
the algal standing crop is determined by different growth rates
which either increase or decrease the standing crop (Figure 1),
addition of nitrogen would not cause an increase in algal concen-
tration unless the addition of nitrogen reduces the rates of the
decreasing factors.
Second, the importance of nutrients other than N, P, and Fe is
illustrated by these findings. The prime importance attached
to nutrients such as N and P should not result in overlooking
other important growth-limiting nutrients or factors such as
micronutrients.
/s ^
Third, the different patterns observed with ^ and X measure^-
ments upon spiking of nutrients illustrates the importance of
measurements. It is not necessary to contend that measure-
ments are more important than it measurements, but the ability
of these measurements to indicate toxic and other factors which
affect growth rates and which are not reflected in X values
warrant their use and further investigation.
Present Status of Batch Culture Assays
The results of this investigation, the previous report [34], and reports
from other evaluation laboratories [55—77] have shown that the batch
122
-------
culture assay, as presently constituted and presented in this report,
is an adequate algal assay in some respects but not in others. For
instance, the algal responses X and ja-^ can be estimated fairly
precisely (Cv ^ 15%) in synthetic media (also between laboratories)
and in natural water samples. Therefore, it is possible to determine
the amount of algal cells that can be produced under optimum condi-
tions in a specific water and also the rate at which they are produced.
Furthermore, the algal response in terms of X, providing that the
necessary precautions are met, appears to be related linearly to the
concentration of the limiting nutrient present, thereby allowing the
determination of the relationship between the addition of a wastewater
(such as secondary effluent) and the maximal amount of algal cells, X,
that can be produced in the laboratory. The batch culture assay also
has the advantages of being relatively simple to perform and is fairly
economical. It seems to be adequate for assaying waters of moderately
to highly enriched nutrient content and toxicity of wastes can be
determined if (1^ measurements are made.
The less satisfactory aspects of the batch assay are centered around
both theoretical and practical concepts. Batch cultures by virtue of
their inherent transient unbalanced growth characteristics with
resultingly wide variations in nutrient and biomass levels, are not
ideal for growth studies where these involve the definition of nutrient-
growth rate relationships. For example, the use of batch culture data
to determine algal growth kinetics using different growth models was
unsatisfactory [34]. The identification of growth-limiting nutrients
by batch assays is still in an unsatisfactory state. The use of yield
coefficients appears to be limited by the gross variations in these
coefficients while spiking (enrichment) techniques are not developed
to the extent where a particular method can be recommended. However,
through further research a satisfactory method will probably be
developed.
One of the greatest limitations of the batch culture at present is the
absence of a clear rational framework in which assay results may be
applied in solving eutrophication problems. Merely obtaining the X
and responses of a test algal species in a natural water sample or
knowing that the addition of a wastewater will increase this & or
still in no way indicates what will happen specifically in a natural
system even for the same algal species. Many other factors influence
the algal standing crop and all of these are reflected by various rates.
To extrapolate batch assay results to real eutrophication problems will
necessitate knowing the identity of the growth-limiting nutrient, the
growth kinetics of various algae for this nutrient, the interaction
between environmental factors and nutrients in regulating algal concen-
trations in nature, etc. For instance, once the identity of the growth-
limiting nutrient has been established the decision on the amount of the
nutrient, if any, that must be removed can only be decided on the basis
of the growth kinetics of algae for this nutrient, i. e. , how does the
nutrient regulate the algal growth rate, how much algal mass is
123
-------
produced per unit mass of the nutrient, etc. The development of the
rational framework for the application of batch bioassay results must
be developed through research on algal growth kinetics and the inter-
relationships of various natural rate factors that determine the algal
standing crop.
The present status of the batch assay is only that of a screening device
for monitoring the general ability of waters to support the growth of
algae and how this ability is grossly affected by wastes, nutrients, or
other growth factors.
CONTINUOUS CULTURE STUDIES
Continuous Culture Growth Kinetics Under
Phosphorus Growth Limitations
Twelve chemostats of the type shown in Figure 16 were used for
continuous culture studies. They were operated under a phosphorus-
limiting condition (i. e. , constant cultural environment and all nutrients
other than phosphate in excess) in a constant temperature room (24°C ±
1°C). Five different residence times were used for each chemostat
(0 = 0, 7 to oo days) during the study period (55 days). Table XVIII
summarizes the conditions for the chemostat assays showing the type
of feed media, the N/P ratio of the feed, the phosphorus content of the
feed (Sc), the residence time, and the number of replicate chemostats.
Phosphorus as the Limiting Nutrient. In the PAAP medium (Tables I
and II), nitrogen is considered to be in excess because the N/P ratio
is 22. 6, McGauhey et al. [30] reported that phosphorus usually was
the rate-limiting nutrient when the N/P ratio was much greater than
15,
Porcella et al. [34] reported that phosphorus was probably the limiting
nutrient in the PAAP medium from chemostat studies, and their batch
results indicated that phosphorus was the only limiting nutrient at a
N/P ratio 2: 40. In this chemostat study the phosphorus content of the
medium was reduced to effect N/P ratios between 40 and 115. This
was done to ensure that phosphorus was the sole limiting nutrient. The
linear relationships between steady state biomass (suspended solids)
and phosphorus uptake at different residence times as shown in Figure
40 indicates that phosphorus was the only limiting nutrient.
Steady State Consideration. The changes in biomass (absorbance) with
time for all 12 chemostats are shown in Figures 41, 42, 43, and 44,
along with input phosphorus concentrations and operating hydraulic
residence times, (The operational data from the chemostat studies
124
-------
TABLE XVIII
CONDITIONS FOR CHEMOSTAT ASSAYS
Medium
Average
N/P Ratio
Average
Residence
Time,
days
No. of
Chemostats
50% PAAP
medium with
S0p = 66 jjLg/I
as P
110
69. 3 & 111
102
104. 8
105
0. 878
1. 31
2. 82
4. 77
GO
3
50% PAAP
medium with
sGp = 100 ng/l
as P
72. 5
59. 5 & 71. 9
68. 2
67. 7
67. 7
0. 928
1. 34
2. 88
4. 97
00
3
50% PAAP
medium with
Sop = 135 \ig/1
as P
54.4
52. 7
52. 0
51. 2
51. 2
0. 889
1. 33
2. 72
4. 81
CO
3
50% PAAP
medium with
sop = 170
as P
43.. 1
41. 8
41.4
40. 8
40. 8
0. 877
1. 31
2. 70
5. 12
00
3
125
-------
140 —
in
¦g
o
a.
tn
3
m
o»
E
<1
ac
t-
z
UJ
o
z
o
o
u
o
UJ
tn
v
o
Ui
t-
V)
60 80 100 120 140
PHOSPHORUS UPTAKE , S0-S, ,jug/Jl as P
FIGURE 40. RELATIONSHIP BETWEEN CELL BIOMASS AND PHOSPHORUS UPTAKE FOR
DIFFERENT RESIDENCE TIMES
-------
.40
I O
0.35 -
2 A
0 30
^3DSc
-W
1.39 , 1.37
3.04
2.91
0951
0.902
0.756
4.89 (4.78, NON-FLOW
1
100 | 62
68.7
68.7
63.3
63.3
66.7
1 1
69.0
1.39 i 'l.32
2.87
2.86
0.940
0.88
0527*
4.69 |4.78, NON-FLOW
1
103 | 64
69 3
69 3
63.3
63.3
66.7
1 1
66.3
1 1.28 , 1.13
2.61
264
0.823
0.773
0.737
4.80 |469| NON-FLOW
1 1
1 1
u
c
¦o
o
ui
O
z
<
CD
ac
o
co
CO
-------
0.40
r\j
OD
40
S.
0.35
Q30
._ 0.25
L ©
s.
V
Is*-
e
1 I 1
116 , 99
1
104
1
104
1
lOO
100
I
lOO
1
103
1
103
1
1.37 , 1.29
3.06
2.87
0955
0925
0.556
482
48S NON-FLOW
1
120 , 99
104
104
93.3
93.3
100
106
1 -
106
—r ¦— -
1.43 , 1.36
3.06
286
0.967
0.861
0.740
5.1
5.03i NON-FLOW
1
117 , 94
100
100
96.3
963
100
101
1 _
101
1
138 , 1.19
2.69
2 74
0.910 |o.869| 0 668
5.08
4.97, NON-FLOW
1
1 1
1
Sampling day C
p.q/1 os P
days
/j.q/1 as P
days
fiq/l as P
days
25 30
TIME, days
FIGURE 42. OPERATION OF CHEMOSTATS 4, 5, AND 6
-------
tNJ
NO
0.4O
035
7 oS°H
0
|9n|0^
1
136
1
. 129
1
135
1
128
1
138
1 1 1
137
1.48 ,
1
1 27
2 74
. 2 88
0.960
,0906
0726
5 08
A93,
N ON-FLOW
1
136
| 130
I
136
130
138
1 1
137
1.45 j
1.20
2.69
. 2 68
0.873
i0.876
0655
4 58
|4.55|
NON-FLOW
137
t 129
133
128
138
136
1 1.30 .
1.25
2.68
i 2.63
0870
,0829
0.739
4.89
CD
NON-FLOW
1
1
QQ
20 25
TIME, days
FIGURE 43. OPERATION OF CHEMOSTATS 7, 8, AND 9
-------
0.05 -
^ /uq/i as P
days
fiq/l as P
days
25 30
TIME, days
FIGURE 44. OPERATION OF CHEMOSTATS 10, 11, AND 12
-------
are tabulated in Appendix D. ) Reasonable steady states were approached
at different phosphorus input levels and different hydraulic residence
times. These operation curves show clearly that the steady state bio-
mass level varies not only with the input phosphorus concentration but
also with the hydraulic residence time (mean cell age, or growth rate).
Figure 40 also indicates that the net yield coefficient (i. e. , Yn =
Xi/(S0-Sj) or slopes in Figure 40) is some function of residence time
instead of a constant coefficient as discussed previously in the theory
section.
Attempts were made to keep the chemostat cultures in a pure culture
condition throughout the whole study period. Microscopic observations
were carried out frequently to check the purity of the cultures. Two
chemostats (Chemostat #9 and #12) were found to be contaminated with
a very small population of bacteria during the later study period
(Chemostat Run 1-9 and 1-12). However, no significant effect on steady
state cell concentration and limiting-nutrient concentration in chemostat
was observed.
Precision of Chemostat Assays. All chemostat assays were done with
three replicate culture systems. The overall precision based on
suspended solids and orthophosphate were estimated and reported in
Table XIX. The suspended solids measurement is more precise than
orthophosphate analysis (Table III), especially at very low orthophosphate
concentration (0.45 to 6. 0 (ig P/i). Consequently, the variance due
solely to the chemostat assay procedure based upon orthophosphate is
higher than that based upon suspended solids. A coefficient of variation
of 6. 96% for the chemostat assay procedure was obtained with respect
to suspended solids measurements, Xj, while an average of 13. 2% was
obtained based on orthophosphate analysis (Si). It appears than an
average of coefficient of variation of about 10% was observed for the
chemostat assay compared to about 15% of batch bioassays reported by
Porcella et al. [34],
Summary of Chemostat Results. The results obtained with the
chemostat experiments in which phosphorus alone constituted the
growth-limiting nutrient are summarized in Tables XX through
XXIV. All data used for kinetic analysis were taken at times when
a steady state was reached (shown with arrows in Figures 41, 42,
43, and 44). Included in Tables XX through XXIV are the steady
state cell concentration (Xj), the hydraulic residence time (0), the
concentration of phosphorus in the influent (SQ) and in the effluent
(Si), the calculated net yield coefficient (Yn = Xl/(S0-Sj)),the
fraction of phosphorus excessively taken up, (fe), the nutrient
removal velocity, (q = (SQ-Sl)/(X1Q))I the amount of excess nutrient
uptake, Se, and the cell phosphorus constant, A.
131
-------
TABLE XIX
PRECISION OF CHEMOSTAT BIOASSAY PROCEDURE
Parameter
Overall Precision
of
dL
Chemostat Bioassays
Precision of
Analytical
Methods
Precision of
Chemostat
Procedure
No. of
Cases
X 1
O
Mean
C- c
V
o
X g
m
Mean
C— d
V
m
c 6
v
a
Suspended
Solids
mg/f
120
37. 0
6. 99
16. 5
1. 30
6. 96
po4-p
\xg/(. P
48
4. 65
13. 3
5. 04
7. 62
10. 40
72
0. 853
23. 0
0. 675
20. 9
16. 0
Si
Variance of chemostat assay (overall) = variance of method plus
variance of chemostat test technique.
^Data taken from Table III.
Q
Cv = coefficients of variation (overall) determined from triplicate
o
chemostats.
C = coefficients of variation determined from the precision of
v
m
methods (Table III).
C = the average coefficient of variation contributed only by the
a
chemostat procedure as expressed by:
I X 2
C =-Jc 2 - C 2 -=22.
V V V V
a F o m o
f_
Xq = arithmetic mean of the results bioassay (overall) response.
®X = arithmetic means of results of analytical methods,
m
132
-------
TABLE XX
SUMMARY OF CHEMOSTAT RUNS A AND B
VjJ
VJJ
P04-P
X,
S -S,
S -S
Chemostat
ec
-"¦l
p J 1
c*
CJ C
v x,
A - 0
0 1
s
f
Run No.
Cell Age
Suspended
Solids
s
0
Si
b -b_
0 1
n ~ S -S,
0 1
A" X,
q " Xj e
e
e
days
1 a n
. _ sv / A 11
mg-cell
Hg P
us P
MO / 0 "P
Fraction
mg /1
M-gIt P
M-g /* p
jj.g/e f
mg P
mg-cell
mg-cell-day
h"-6' 1
A - 1
0. 951
11. 1
64. 8
4. 80
60
185
5. 41
5. 68
46. 2
0. 770
A - 2
0. 940
11. 7
63. 3
5. 10
58. 2
201
4. 97
5. 29
43. 7
0. 751
A - 3
0. 823
11.4
63. 3
5.25
58. 0
197
5. 09
6. 18
43. 8
0. 755
A - 4
0. 955
15. 7
100. 0
5.40
94. 6
166
6. 03
6. 31
75. 1
0. 796
A - 5
0. 967
13. 5
93. 3
4. 80
88. 5
153
6. 56
6. 78
71. 7
0. 810
A - 6
0. 910
14. 7
96. 3
5. 1
91. 2
161
6. 20
6. 82
72. 9
0. 799
A - 7
0. 960
19. 2
128. 0
4. 95
123
156
6.41
6. 67
99. 0
0. 805
A - 8
0. 873
20. 1
130. 0
4. 65
125
161
6. 22
7. 12
100. 0
0. 800
A - 9
0. 870
18. 3
128. 0
5. 10
123
149
6. 72
7. 73
100. 2
0. 815
A - 10
0. 867
23. 8
161. 0
5.85
155
154
6. 51
7. 51
125. 4
0. 809
A - 11
0. 924
25. 8
162. 0
5. 70
156
165
6. 05
6. 54
124. 0
0. 795
A - 12
0. 920
27. 5
164
6.0
158
174
5. 75
6. 25
123. 9
0. 784
B - 1
0. 902
10. 0
64. 8
5. 1
59. 7
168
5. 97
6. 62
47. 2
0. 791
B - 2
0. 880
12. 9
63.3
5. 1
58. 2
222
4. 51
5. 13
42. 2
0. 725
B - 3
0. 773
11. 2
63. 3
4.95
58. 3
192
5. 20
6. 73
44. 4
0. 762
B - 4
0. 905
13. 5
100
5. 1
94. 9
142
7. 03
7. 77
78. 1
0. 823
B - 5
0. 861
14.4
93. 3
5.4
87. 9
164
6. 10
7. 09
70. 0
0. 796
B - 6
0. 869
13. 9
96. 3
5. 25
91.0
153
6. 54
7. 53
73. 7
0. 810
B - 7
0. 926
21. 1
128
5. 25
123
172
5. 83
6. 30
96. 8
0. 787
B - 8
0. 876
20. 0
130
5.4
125
160
6. 25
7. 13
100. 1
0. 801
B - 9
0. 829
20. 3
128
5. 25
123
165
6. 06
7. 31
97. 8
0. 795
B - 10
0. 817
26.4
161
5. 85
155
170
5. 87
7. 19
122. 1
0. 788
B - 11
0. 865
27.4
162
6. 45
155
170
5. 66
6. 54
121. 1
0. 781
B - 12
0. 866
27. 6
164
6. 3
158
175
5. 72
6. 61
123. 7
0. 783
-------
TABLE XXI
SUMMARY OF CHEMOSTAT RUNS C AND D
H*
V-M
-p-
Chemostat
Run No.
0c
Cell Age
days
X!
Suspended
Solids
mg It
PCU-P
S
O
Si
S -S,
0 1
Y - X'
S -S,
a 0 1
S -S,
a 0
S
e
f
e
n ~ S -S,
0 1
" x,
q " xxe
Kg It P
^gIt P
fJLgIt P
mg-cell
Wt P
UK P
fig/f P
Fraction
mg P
mg-cell
mg-cell-day
C - 1
C - 2
C - 3
C - 4
C - 5
C - 6
C - 7
C - 8
C - 9
C - 10
C - 11
C - 12
1. 39
1. 39
1. 28
1. 37
1.43
1. 38
1.48
1.45
1. 30
1. 32
1. 35
1.45
18. 8
19. 5
18. 6
25. 0
27. 2
24. 6
33. 1
29. 1
27. 0
41. 2
37. 9
35. 9
100
100
103
116
120
117
136
136
137
170
171
175
3. 3
4. 5
3. 3
3.6
3. 9
3. 9
3. 0
4. 5
3. 9
4.8
4. 2
5.4
96. 7
95. 5
96. 7
112
116
113
133
132
133
165
167
170
194
204
192
223
234
218
249
221
203
250
227
211
5. 14
4. 89
5. 20
4. 48
4. 26
4. 59
4. 02
4. 54
4. 93
4. 00
4. 41
4. 74
3. 70
3. 52
4. 06
3. 27
2. 98
3. 33
2. 71
3. 13
3. 79
3. 03
3. 26
3. 27
73. 3
71. 2
73. 6
80. 9
82. 2
82. 4
91. 9
95. 8
99.4
113. 9
119. 9
125. 3
0. 758
0. 746
0. 761
0. 722
0. 709
0. 729
0. 691
0. 726
0. 747
0. 690
0. 718
0. 737
D - 1
1. 37
13. 5
63
3. 6
59.4
227
4.40
3. 21
42. 6
0. 717
D - 2
1. 32
16. 6
62
3. 0
59. 0
281
3. 55
2. 69
38. 4
0. 651
D - 3
1. 13
11. 7
64
3. 3
60. 7
193
5. 19
4. 59
46. 2
0. 761
D - 4
1. 29
20. 7
99
3. 0
96. 0
216
4. 64
3. 60
70. 3
0. 732
D - 5
1. 36
24. 3
99
3. 3
95. 7
254
3. 93
2. 89
65.4
0. 684
D - 6
1. 19
21. 9
94
3. 6
90. 4
242
4. 13
3. 47
63. 2
0. 699
D - 7
1. 27
27.4
129
3. 9
125
219
4. 56
3. 59
91. 0
0. 728
D - 8
1. 20
32. 1
130
4. 5
125
257
3. 89
3. 25
85. 1
0. 681
D - 9
1. 25
30. 3
129
4. 2
125
242
4. 13
3. 30
87. 4
0. 699
D - 10
1. 22
39. 3
161
5. 25
156
252
3. 97
3. 25
107. 2
0. 687
D - 11
1. 22
37. 5
163
4. 35
159
236
4. 24
3. 48
112.4
0. 707
D - 12
1. 32
41. 5
165
4. 95
160
259
3. 86
2. 92
108. 5
0. 678
-------
TABLE XXII
SUMMARY OF CHEMOSTAT RUNS E AND F
v/i
Chemostat
Run No.
ec
Cell Age
days
x,
Suspended
Solids
mg/f
po4-p
S
o
S
l
S -S,
O 1
v X,
S -S,
o 1
S -S,
O 1
S
e
f
e
n ~ S -S,
o 1
X,
q ~ xx e
y-g/t P
(Xg/f P
[Lg/l P
mg-cell
Hg P
ng p
^g/? P
Fraction
mg P
mg-cell
mg-cell-day
E - 1
E - 2
E - 3
E - 4
E - 5
E - 6
E - 7
E - 8
E - 9
E - 10
E - 11
E - 12
3. 04
2. 87
2. 61
3. 06
3. 06
2. 69
2. 74
2. 69
2. 68
2. 62
2. 68
2. 86
22.4
24. 7
20. 1
32. 2
37. 9
36. 1
51. 6
54. 2
51. 5
66.3
66.9
61. 5
67. 5
68. 7
69. 3
104
104
100
135
136
133
166
172
169
1. 2
1. 2
1. 35
1. 05
1. 2
1. 35
1. 35
1. 5
1. 2
1. 65
1. 65
1. 8
66. 3
67. 5
68
103
103
98. 6
134
134
132
164
170
167
338
366
296
314
369
366
385
404
390
404
394
380
2. 96
2. 73
3. 38
3. 20
2. 72
2. 73
2. 60
2. 47
2. 56
2.47
2. 54
2. 72
0. 974
0. 952
1. 30
1. 05
0.89
1. 02
0. 95
0. 92
0. 96
0. 94
0. 95
0. 95
38. 5
36. 8
43. 0
63. 0
55. 9
53. 6
69. 9
67. 0
67. 9
81. 5
86. 9
90. 5
0. 580
0. 545
0. 632
0. 612
0. 543
0. 544
0. 522
0. 500
0. 515
0.497
0. 511
0. 542
F - 1
2. 91
20. 3
67. 5
1. 5
66. 0
308
3. 25
1. 12
40. 8
0. 618
F - 2
2. 86
20. 0
68. 7
1. 35
67. 3
297
3. 37
1. 18
42. 5
0. 631
F - 3
2. 64
21. 0
69. 3
1. 5
67. 8
310
3. 23
1. 22
41. 7
0. 615
F - 4
2. 87
29. 6
104
1. 2
103
288
3. 48
1. 22
66. 2
0. 643
F - 5
2. 86
29. 9
104
1. 5
102
292
3. 41
1. 19
64. 8
0. 635
F - 6
2. 74
28. 8
100
1. 5
98. 5
291
3.42
1. 25
62. 6
0. 636
F - 7
2. 88
45. 3
135
1. 5
133
337
2. 93
1. 02
76. 7
0. 577
F - 8
2. 68
42. 0
136
1. 65
134
313
3. 19
1. 19
81. 9
0. 611
F - 9
2. 63
45. 2
133
1. 65
131
345
2. 89
1. 10
74. 8
0. 571
F - 10
2. 61
58. 2
166
1.8
164
355
2. 81
1. 08
91. 7
0. 559
F - 11
2. 58
59. 5
172
1. 35
171
348
2. 87
1. 11
97. 1
0. 568
F - 12
2. 82
57. 1
169
1. 65
167
338
2. 92
1. 04
96. 0
0. 575
-------
TABLE XXIII
SUMMARY OF CHEMOSTAT RUNS G AND H
CTn
Chemostat
Run No.
®c
Cell Age
days
x*
Suspended
Solids
mg/e
P04-P
S
O
S!
S -S
0 1
Y X,
S -S,
\ 0
S -S,
0 1
S
e
f
e
n S -S
0 1
A" Xl
q ~ xx e
M-g/f P
|Xg/f P
H-g/f P
mg-cell
H-g P
ng p
tig/e P
F raction
mg P
mg-cell
mg-cell-day
G - 1
G - 2
G - 3
G - 4
G - 5
G - 6
G - 7
G - 8
G - 9
G - 10
G - 11
G - 12
4. 89
4. 69
4. 80
4. 82
5. 10
5. 08
5. 08
4. 58
4. 89
4. 78
5. 18
5. 51
30. 9
32. 7
32. 2
55. 4
54. 8
55. 8
72. 8
75. 4
71. 4
91. 5
91. 2
94. 7
65. 1
69. 0
66. 3
103
106
101
137
137
136
172
172
171
0. 6
0. 45
0.45
0. 45
0. 60
0. 9
0. 7 5
0. 6
0. 9
0. 75
0. 75
0.9
66. 5
68. 5
65. 8
102
105
100
136
136
135
171
171
170
479
477
489
543
522
558
535
554
529
535
533
557
2. 09
2. 09
2. 04
1. 84
1. 92
1. 79
1. 87
1. 80
1. 89
1. 87
1. 88
1. 79
0. 43
0. 45
0. 426
0. 382
0. 376
0. 353
0. 368
0. 394
0. 387
0. 391
0. 362
0. 326
26. 1
27. 9
25. 8
33. 2
36. 9
30. 6
45. 6
42. 8
46. 2
57. 5
57. 8
52. 4
0.405
0. 407
0. 392
0. 325
0. 351
0. 306
0. 335
0. 315
0. 342
0. 336
0. 338
0. 308
H - 1
4. 78
29. 9
65. 1
0. 6
64. 5
464
2. 16
0. 451
27. 3
0. 423
H - 2
4. 78
31. 3
69. 0
0.45
68. 5
457
2. 19
0. 458
29. 6
0. 432
H - 3
4. 68
30. 8
66. 3
0. 6
65. 7
469
2. 13
0. 456
27. 4
0. 417
H - 4
4. 82
51. 8
103
0. 75
102
508
1. 97
0. 409
37. 5
0. 368
H - 5
5. 03
51. 2
106
0. 7 5
105
488
2. 05
0. 408
41. 4
0. 394
H - 6
4. 97
49. 1
101
0. 75
100
491
2. 04
0. 410
39. 0
0. 390
H - 7
4. 93
70. 7
137
0. 75
136
520
1. 92
0. 390
48. 3
0. 355
H - 8
4. 55
69. 5
137
0. 75
136
511
1. 96
0. 430
49. 6
0. 365
H - 9
4. 85
70. 2
136
0. 90
135
520
1. 92
0. 396
47. 8
0. 3 54
H - 10
4. 62
89. 0
172
0. 9
171
520
1. 92
0. 416
60. 4
0. 3 53
H - 11
5. 18
88. 5
172
0. 75
171
518
1. 93
0. 373
61. 0
0. 3 57
H - 12
5. 42
87. 7
171
0. 9
170
516
1. 94
0. 357
60. 9
0. 3 58
-------
TABLE XXIV
SUMMARY OF CHEMOSTAT RUNS I AND J
Chemostat
Run No.
ec
Cell Age
days
^max
Suspended
Solids
mg/f
PO4-P
Hg /« P
jig n p
s -s
o 1
\Lgll P
max
max
S -S,
o 1
mg-cell
mg P
A =
o
S -S,
o 1
M>g P
mg-cell
- 1
- 2
- 3
- 4
- 5
- 6
- 7
- 8
- 9
- 10
- 11
- 12
Nonflow
System
0C = CO
Yn = Ymax
q = 0
49. 6
52. 3
48. 0
85. 1
84. 2
86. 0
120
107
111
141
131
141
65. 1
69. 0
66. 3
103
106
101
137
137
136
172
172
171
0. 15
0. 15
0. 3
0. 3
0.3
0.45
0.45
0. 3
0.45
0.45
0.45
0. 45
64.
9
68.
8
66.
0
103
106
100
136
137
135
171
171
170
64.
8
68.
7
65.
7
103
105
100
136
136
135
171
171
170
764
760
727
826
794
860
883
779
824
826
768
827
1. 31
1. 32
1. 38
1. 21
1. 26
1. 16
1. 13
1. 28
1. 22
1. 21
1. 31
1. 21
J -
J -
J -
J -
J -
J -
J -
J - 8
J - 9
J - 10
J - 11
J - 12
Nonflow
System
0C = 00
^n ~ ^max
q = 0
Se = 0
50. 8
54. 7
47. 4
84. 2
84. 3
85. 0
119
111
111
130
130
131
65. 1
69.0
66. 3
103
106
101
137
137
136
172
172
171
0. 3
0. 3
0. 6
0. 3
0.45
0.45
0. 6
0.45
0.45
0. 6
0. 6
0. 6
784
796
721
817
803
850
876
816
823
760
766
772
1. 28
1. 26
1. 39
1. 22
1. 25
1. 18
1. 14
1. 23
1. 22
1. 32
1. 32
1. 30
-------
Kinetic Analyses and Kinetic Constants. The growth character-
istics of Selenastrum capricornutum were evaluated in terms of
the following kinetic analyses, and kinetic constants were obtained
with respect to a phosphorus-limiting condition:
a. The net yield coefficient (Yn) and its relation to cell age, and
nutrient utilization constants, K^\, Kg.
Y„ = c = Y I 1 - e (39)
and
Y Y 6
v - x _ max c .. 7.
n " S - S, " 0 + K„ * { '
o i c B
b. Nutrient removal velocity and nutrient utilization rate for
cell growth.
c. The maximum specific growth rate, p., (day-1).
d. The decay rate, k^, (day-1).
e. The half saturation constant, Kg, (|ig P/£).
Net Yield Coefficient and Cell Continuity Equation. The net yield
coefficients calculated from steady state cell concentrations and
phosphorus uptake are listed in Tables XX to XXIV. The yield
coefficients were significantly affected by the hydraulic residence
time (or growth rate). The net yield increased with cell age up
to a maximum cell yield coefficient (Ymax). As discussed in the
theory section (Section IV) this phenomenon appears to be caused
primarily by the so-called "excess" nutrient uptake or more
precisely "excess" phosphorus uptake. The cell continuity
equations (Equation 20) modified to include a variable yield coef-
ficient can be simplified and reduced to the following net yield
coefficient-hydraulic residence time equations:
/ "6c/KA
Y = Y 1 - e (39)
n max I '
and
Yn ' Ymax ( 6c !CKb ' ' (42»
138
-------
The maximum cell yield coefficient is defined as the cell yield
coefficient at which there is no "excess" phosphorus uptake and
it is determined experimentally in a nonflow cultural system
(i. e. , 0 -* oo days).
In a nonflow continuous culture system, the growth rate of the
organism will approach the decay rate, i. e. , jjl = 1/0 + -»¦
when 0 —»¦ co. The cell concentration will thus be maintained at a
constant value where the cell matter produced per unit amount of
nutrient utilized approaches the maximum yield coefficient. Thus:
X
Y = c ma^ , when 0^ oo, Xj -X
max S - S 1 max
o 1
A plot of maximum steady state biomass (Xmax) against phosphorus
uptake is shown in Figure 45. A maximum cell yield coefficient
of 805 mg cell mass produced per mg of phosphorus removed was
obtained with a correlation coefficient r = 0. 99 (p > 0. 99). This
value corresponded to a cellular constant of 0. 124% phosphorus.
By direct nonlinear and indirect linear transformation least square
regression analyses of Equation 39 and 42, the best'fit equations
are (Figure 46):
Yn = s *'s = 805 1 1 - e 4;78j (68)
and
v 805 0
Y = ——1 = (69)
n S - S, 0_: +' 2. 98 [°^}
o X c
in which = 4. 78 days and Kg = 2. 98 days. Although both
equations are highly significant (p < 0. 01), the exponential
equation (Equation 68) has a better correlation coefficient
(r = 0. 99) compared to r = 0. 98 for the rectangular hyperbolic
equation (Equation 69). The nutrient utilization constant, K^,
is estimated to be 4. 78 days residence time which corresponded
to a net yield, Y = 0. 632 (maximum cell yield coefficient) (i. e. ,
Yn = 0. 632 Ymax) while the nutrient utilization constant Kg is
estimated to be 2. 98 days residence time which corresponds to
a net yield, Yn = 0. 5 (maximum cell yield coefficient) (i. e. ,
Yn = 0. 5 Ymax).
139
-------
FIGURE 45. RELATIONSHIP BETWEEN STEADY STATE CELL CONCENTRATION
AND PHOSPHATE UPTAKE AT INFINITE HYDRAULIC RESIDENCE
TIME
-------
FIGURE 46. NET CELL YIELD COEFFICIENT AS A FUNCTION OF MEAN CELL
AGE, 0
c
-------
Nutrient Removal Velocity and Excess Phosphate Uptake.
As discussed in the theory section, when the nutrient removal
velocity is greater than the rate of nutrient utilization for cell
growth, the nutrient is accumulated and stored inside the cell
in excess of the minimum nutrient requirement, or in other
words "excess" nutrient uptake occurs.
The phosphorus removal velocities, q, at different cell ages are
plotted against the specific growth rates, jjl, in Figure 47. When
no excess phosphorus uptake occurs, the equilibrium between the
phosphorus removal velocity, q, and the phosphorus utilization
rate, n/Ym_x, can be described by equation q = Ymax as shown
by the dottealine in Figure 47. Consequently, no excess of phos-
phorus accumulates inside cells. On the other hand, the equation
q = jjl/ (Yn (1 + 0C kj)) indicates the inequality of the phosphorus
uptake rate and utilization rate, and the effect of phosphorus
accumulation and storage. The shaded area in Figure 47 represents
the degree of excess phosphorus uptake rate with no excess
phosphorus uptake at jjl = k days and any phosphorus content higher than 0. 124%
is considered to be in excess, Equations 4 5 and 46 were developed.
Figure 48 shows the variation in the fraction of excess phosphorus
uptake, Se/(S0.Sj), with hydraulic residence time.
0
c
f = e 4* 78 (r = 0. 00, p > 0. 99)
(70)
or
fe ~ 0 + 2. 98
c
e
c
(r = 0. 98, p > 0. 99) .
(71)
142
-------
SPECIFIC GROWTH RATE,/^= -g- + kd ,day-
FIGURE 47. COMPARISON OF NUTRIENT REMOVAL VELOCITY,
q, AND SPECIFIC GROWTH RATE, jjl
143
-------
Again according to the correlation coefficient, Equation 70 gives
a better fit to the experimental data than Equation 71. Figure 48
indicates that the fraction of excess phosphorus uptake, Se/(S0-S1),
is a simple function of mean cell age in the chemostat and it
appears to be independent of input phosphorus concentration.
"Washout" Residence Time and Maximum Specific Growth Rate. In
chemostat operation without cell recycle, the specific growth rate
of algal cells is regulated by the hydraulic residence time (0 = 9C)
and is related to hydraulic residence time by,
where k
-------
Ul
4 5 6 7
CELL AGE, 0C ,doys
FIGURE 48. FRACTION OF "EXCESS" PHOSPHORUS UPTAKE AS FUNCTION OF
MEAN CELL AGE, e
c
-------
Decay Constant, k^. A decay rate of 0. 001 per day was obtained
directly from Equation 72 by nonlinear regression analyses. The
small magnitude of k^ indicates that the rate of decrease in cellular
mass due to endogenous respiration and cellular death with sub-
sequent lysis is very small compared to the cell growth rate.
Unfortunately many investigators neglect or ignore the decay rate
because of its small magnitude. However, the decay rate which
may be as large as 0. 05 day-1 can play an important role in
kinetic analyses. The small magnitude of the decay rate in this
study (phosphorus-limiting condition) might be caused by the
nutrient (phosphorus) "excess" uptake. Due to the high rate of
nutrient uptake, the nutrient (phosphorus) released by cellular
lysis may be taken up by cells immediately for cellular growth.
Maximum Specific Growth Rate. The maximum specific growth
can be estimated from Equation 30 (1/0W = + k^) by knowing the
"washout" residence time and decay constant. A maximum growth
rate of 1. 85 per day was obtained for Selenastrum capricornutum
under the specific phosphorus-limiting condition. This value is
in the same order of magnitude reported by the PAAP research
group [34, 68]. This is no doubt but that the maximum specific
growth should be greater than (1 /Grnin + ^d) where 6min is the
minimum hydraulic residence time in which cells can maintain
their steady state population without being washed out. In chemostat
run #4, 9min was found to be 0. 556 day which corresponds to a
specific growth rate of 1. 80 day"1.
Half Saturation Constant, Ks. A half saturation constant of 5. 67
|j.g P/£ was obtained directly from Equation 72 by nonlinear
regression analysis as shown by curve A in Figure 49. However,
the best fit linear transformation (Sx = Ks \il (pi - jjl); Sx as the
dependent variable) gives a half saturation constant of 3. 72 jxg Pft
with a correlation coefficient of 0. 945 as shown by curve B in
Figure 49. It is difficult to assign a single value for the saturation
constant, Ks; however, it appears that a range from 3. 7 to 5. 7
Hg P11 best represents the most probable values for Kg of
Selenastrum capricornutum with respect to a phosphorus-limiting
condition. These low values for the half saturation constant for
phosphorus imply that if phosphorus has been identified as the
rate-limiting nutrient in Lake "A" and if the phosphorus concen-
tration in Lake "A" is about 10 (Jig PIt, a high rate of algal growth
(Figure 49 shows a growth rate from 1. 2 day"1 to 1. 5 day"1 for
Selenastrum capricornutum at a phosphorus concentration of
10 |ag P/£ ) likely exists in Lake "A. "
146
-------
2.0
^ * i
1.85 day"'
o
"O
lii
t-
<
C£
5
o
cr
o
v
UJ
a.
-------
Summary of Phosphorus Limitation Chemostat Studies
The most probable kinetic growth coefficients and constants developed
from the data analysis for Selenastrum capricornutum under a
phosphorus-limiting condition are summarized in Table XXV. Table
XXV includes the coefficients and constants for all models evaluated
and the results for each computational method. It is noted that all
the regression equations in Table XXV are at a significance level of
less than 1. 0% (i. e. , p £ 0. 99).
Considering both the conceptual framework for the growth relation-
ships as well as the level of statistical significance of the experimental
data conforming to the kinetic models considered, the most probable
models (best analytical formulations) and the best estimate of the
kinetic constants and coefficients are summarized in Table XXVI.
It ma^be noted from Table XXVI that Equation 39 (Yn = Ymax (1 -
e A)) best describes the relationship between net yield coef-
ficient and cell age. The rectangular hyperbolic equation (Yn =
6C/ (Kg + 6C)» Equation 42) can probably also be used to
describe this net yield—cell age relationship but it does not appear
to fit the experimental data as well as the exponential equation.
The half saturation constant, Kg, was found to be between 3. 7 and 5. 7
|JLg P11 for Selenastrum capricornutum under phosphorus-limited
conditions and it appears to be the most significant growth constant
observed in this study. It should be noted that 3—6 |ig P/0 for K is
a very low level and its significance can be outlined with a simple
practical example. Assume that in Lake "B" phosphorus is presumed
to be the rate-limiting nutrient and serious consideration is being
given to the partial, say 80%, removal of phosphorus from waste
inputs which contribute presently about 50% of the total reactive
phosphorus input to the lake.
Case 1. If the phosphorus concentration in Lake "B" is about 5 jxg
P11 ( = Ks of phosphorus for Selanstrum capricornutum), it is at a
level where the growth rate is significantly affected by the phosphorus
concentration (approximately first order). Thus any reduction (say
80%) in the phosphorus content of the waste discharges would reduce
proportionately the concentration of phosphorus in the lake (~40%)
as well as the algal growth rate. Obviously the reduction in phosphorus
concentration of the lake would only be affected by the fraction of the
total phosphorus input removed by waste treatment. However, as the
algal growth rate is almost linearly related to the phosphorus concen-
tration in the lake, the maximum reduction of phosphorus possible
would be beneficial from the standpoint of reducing the algal growth
rate as well as the standing crop of algae.
148
-------
TABLE XXV
KINETIC GROWTH MODELS AND CONSTANTS FOR Selenastrum capricornutum
UNDER PHOSPHORUS LIMITING CONDITIONS
(Various Models and Computational Methods)
Kinetic Model
C omputati onal
Method
Growth Kinetic Constants
and Equations
Correlation
Coefficient
Significance
Level at:
Y . X, ,,
Yn " SQ - Sl max ^ J
y e
max _ c
Y - Y Ka
max n A
Linear
Transformation
Regression
Analysis
f -0 /4. 1S\
Yn = 805 V"e ' /
Qjj, me cells produced
max mg P utilized
Ka = 4. 78 days
0. 99
<1%
y _ y / Bc )
n maxl 0c + Kg J
y e
n _ c
y - y k„
max n 13
Linear
Transformation
Regression
Analysis
Yn- 805 (ec + U8)
xr mg cells produced
max ~ mg P utilized
Kr = 2. 98 days
0. 98
<1%
u. ?S1
s + Si
(?L+kd)Sl
? + kd ~ ks + S,
Nonlinear
Regression
Analysis
1. 85 S,
^ 5. 67 + S,
= 0. 001 day"1
0 = 0. 54 day
w 1
(1 = 1.8 5 day"1
Kfl = 5. 67 ng Pft
<1%
s -hi
- A
\x - n
Linear
Transformation
Regression
Analysis
1, 85 ST
^ 3. 72 + S,
G = 0. 54 day
w '
fx = 1. 85 day"1
Kfl = 3. 72 fxg Pit
0. 95
<1%
A
S, = K In —li—
a M. - t*
Linear
Transformation
Regression
Analysis
l.85(l -e-V6.06j
fx = 1,85 day"1
Ka = 6.06 (ig PIt
Kb = 4. 19 (ig P/f
0. 98
<1%
tt = v-S'"
Nonlinear
Regression
Analysis
,, = 1. 85 S, 1,07
5. 83 + S, 1- 07
(i = 1.8 5 day-1
n = 1. 07
Kg = 5. 83 pg P/t
<1%
1^9
-------
TABLE XXVI
THE MOST PROBABLE KINETIC GROWTH MODELS AND BEST
ESTIMATE OF GROWTH CONSTANTS AND COEFFICIENTS
FOR Selenastrum capricornutum UNDER
PHOSPHORUS LIMITED CONDITIONS
Kinetic Growth Relationship
Most Probable Equation and Best
Estimate of Coefficient
Net yield coefficient - cell age
relationship
(Modified Cell Continuity-
Equation)
Y = X1 = Y (l -
n SQ - Sj max I J
Y - 20r ce^s Pr°duced
max ~ mg P utilized
Ka = 4. 78 day
Fraction of "excess" phosphorus
uptake — cell age relationship
-0 /KA
c C
f = e
e
K^ = 4. 78 day
Specific growth rate — rate
limiting nutrient concentration
relationship
Michaelis-Menten, Monod Equation
U= PS1
" si + k8
k^ = 0, 00 1 day-1
jl = 1. 85 day"1
K = 3. 7 ~ 5. 7 lig P/f
s
150
-------
Case 2. If the phosphorus content in Lake "B" is about 50 |xg P/£
(= 10 x Kg), the specific growth rate of Selenastrum capricornutum
is near its maximum value (i. e, , p 1. 8 5 day-1). Any increase in
the phosphorus concentration of the lake would have no measureable
effect on the maximum growth rate nor on the standing crop.
Correspondingly, a reduction of say 80% of the waste discharge
phosphorus would only reduce the total phosphorus input by 40% and
the phosphorus concentration in the lake would still be about 30 jig
P/I. At this level of phosphorus concentration (i. e. , ~ 6 x Ks), the
specific growth rate is still very near its maximum level (i. e. ,
H -v 1. 80 day"1) and correspondingly there would be no measureable
decrease in the standing crop of algae. The expenditure for phos-
phorus removal would be essentially wasted. It would have been
far more effective if a part of the money spent for phosphorus
removal had been spent on determining the growth rate limiting
nutrient or factors. It is obvious that if the phosphorus content is
as high as 50 fig P/St, it is very unlikely (if not impossible) that
phosphorus is rate-limiting for an organism with the growth
characteristics of Selenastrum capricornutum. Once the growth-
limiting nutrient or factor is identified, attention can be directed
to the economic and technical feasibility of its removal and the
likely effects on the growth rate and standing crop of algae.
Chemostat Study on Natural Water (Sacramento River)
Chemostat algal bioassays were conducted on samples of the
Sacramento River (California) to determine: 1) the linearity of
steady state cell concentration response, 2) the identification of the
growth-limiting nutrient, and 3) the interpretation and application
of kinetic growth constants from natural water chemostat bioassays.
Water samples from the Sacramento River were pretreated by auto-
claving. Twenty-liter aliquots of 100%, 66. 7%, and 33.3% (V/V)
were produced by diluting Sacramento River waters with autoclaved
distilled deionized water. Twelve chemostats .(3v-30% PAAP
control, 3/tr-100% Sacramento River waters, 3 V-66. 6% Sacramento
River water and 3^33. 3% Sacramento River samples) were operated
at a hydraulic residence time of about 2. 6 days. Cell counts by
hemocytometer and pH measurements were performed daily, while
nutrient analyses (NHt-N, (N02 + N03)-N, POf-P), cell counts, and
pH were performed at steady state conditions. Table XXVII sum-
marizes all the daily and steady state measurements for each
parameter.
The Linearity of Steady State Cell Concentration Response. Steady
state cell concentration based on cell counts and suspended solids
were determined and related to the dilutions of Sacramento River
water as shown in Figure 50. A highly significant (p > 0. 99)
151
-------
TABLE XXVII
SUMMARY OF CHEMDSTAT OPERATION RESULTS ON SACRAMENTO R.'VER WATER
Day of Operation
5
6
7
8
Hydraulic
Number
Sample
ODa
OD
OD
OD
po4
-P
nh4
-X
NO, + NO / -N
T im o.
PH
0
Cellb
PH
Cell
PH
Cell
pH
Cell
Solid s
s
S,
S
S,
S
S
Counts
Counts
Counts
Counts
mg/ f
o
^g N/f
o
M-g N/f
ng PIt
P/f
Hg y-n
Hig N/f
fday s)
1
Control
8. 0
0. 283
8. 2
0. 260
8. 3
0. 235
8. 3
0. 225
76. 0
186
1. 5
0
^ 0
4200
628
2.8 5
30% PAAP
2
7. 8
0. 285
8. 2
o
o
8. 3
0. 220
8. 4
0. 195
66. 0
186
1. 2
0
^ 0
4200
579
2. 35
3
7. 9
0. 220
7. 9
0. 200
8. 2
0. 200
8. 4
0. 190
64. 5
186
2. 1
0
0
4200
554
2. 30
4
100%
7. 5
223
8. 0
146
8. 3
162
8. 2
147
18. 7
34. 2
2. 7
25. 2
- 0
107. 6
Q. 8
2. 85
Sacramento
5
River
8. 1
167
7. 5
125
8.4
154
8. 0
150
21. 9
34. 2
3. 9
25. 2
2. 1
107. 6
11. 1
2. 94
Water
6
8. 0
154
7. 8
139
8.4
114
7. 7
135
20. 2
34. 2
4. 2
2 5. 2
- 0
107. 6
8. 2
2. 85
7
66. 7%
8. 1
130
8, 0
106
8. 1
121
7. 5
112
8. 3
22. 1
3. 3
16. 6
0. 42
75. 0
9. 3
2. 56
Sacramt-nto
8
River
8. 1
144
7. 7
100
8. 0
103
8. 1
102
7. 1
22. 1
4. 5
16. 6
^ 0
75. 0
6. 5
2. 56
Water
9
7. 8
143
8. 3
85
8. 0
Q8
7. 9
Q5
6. 1
22. 1
4. 8
16. 6
- 0
7 5. 0
6. 5
2. 4 1
10
33. 3°o
7. 9
37
8.4
35
8. 2
52
8. 0
48
4. 1
9. 7
4. 8
14. 3
0. 84
36. 7
8. 5
2. 63
Sacramento
li
River
8. 1
90
8. 5
79
8. 3
85
7. Q
70
5. 1
9. 7
4. 8
14. 3
0
36. 7
9. 5
2. 60
Water
12
8. 2
39
8. 3
19
8.4
40
7. 9
45
3. 0
9. 7
4. 8
14. 3
0. 84
364 7
11. 3
2. 56
aOD = absorbance at 600 mji.
^Cell counts = cell number x 556/m£ (hemocytomr>ter).
-------
260
-4
33.3 66.6 100
PERCENT OF SACRAMENTO RIVER WATER IN DILUTION
FIGURE 50. RELATIONSHIP BETWEEN STEADY STATE
CELL CONCENTRATION AND DILUTION OF
SACRAMENTO RIVER WATER
153
-------
correlation coefficient (r = 0. 97, n = 9) was observed between the
steady state cell counts (hemocytometer) and the dilutions of
Sacramento River water.
Cell counts (x 556 cells/mi) = 10. 8 + 1. 34 (% Sacramento River Water) .
(73)
On the other hand, although a highly significant (p >0. 99) correlation
coefficient (r = 0. 93, n = 9) was obtained between the steady state
suspended solids and the dilutions of Sacramento River water, this
relationship shown in Figure 50 and described below,
Suspended solids (mgI!) = 0. 243 (% Sacramento River Water) - 5. 7
(74)
was unsatisfactory due to the large scatter in the data. The large
scatter in the data was likely due to the fact that the nutrient level in
diluted Sacramento River samples could not support enough algal
growth to produce a high enough level of Xx (suspended solids = 4 mg/i
@33%) to insure adequate precision with the suspended solids determin-
ation. One of the problems with the use of suspended solids as a cell
concentration parameter is inadequate precision at low biomass levels.
Also, there was considerable suspended particulates in the Sacramento
River water which interfered with the use of a suspended solids
measurement as a cell concentration parameter.
The linearity of steady state cell concentration (hemocytometer cell
counts) and the dilution of natural water indicates that there is no
evidence of toxicants.present in Sacramento River water. It is
readily apparent that the chemostat can be used efficiently for assaying
for relative algal response for waters of moderate nutrient content
as well as for waters rich in nutrients.
Algal Growth Kinetic Characteristics. Table XXVIII summarizes the
observed parameters of cell concentration, soluble orthophosphate
and nitrate-nitrite-ammonia nitrogen for the presumed steady state
conditions for the chemostat assay of Sacramento River water. The
cell concentration expressed in terms of suspended solids was esti-
mated from the relationships between cell counts, absorbance, and
suspended solids reported by Porcella et al. [34],
Net Yield Coefficient ^/.(Sq-S ). The net yield coefficients were
calculated based on phosphate and total soluble nitrogen (NOJ +
NO2 + NH4-N). A high variation of net yield coefficient for
phosphate was observed, the values ranging from 417 to 1310 mg
154
-------
TABLE XXVIII
PRESUMED STEADY STATE CHEMOSTAT PARAMETERS FOR SACRAMENTO RIVER WATER
H
VJ1
vn
X,
b
PO4-P
NO 3
+ NOz + NH,-N
Chemostat
Type of
Biomass
Residence
Specific
Net Yield
Calculated
Net Yield
Calculated
Number
Sample
Suspended
Time
Rate
S
o
S,
X, me cell
K
SD
S
O
S,
X, mg cell
K
(mg/0
(days)
day"1
ng Pit
Pit
S -S mg P
0 1 6
Hg Pit
Hg N/l
(ig N/<
So-Sj mg N
M-g N/f
1
Control
85. 7
2. 85
0. 351
186. 0
1. 5
454
6.4
4200
628
24. 0
2680
30% PAAP
2
74. 2
2. 35
0.426
186. 0
1. 2
401
4. 0
4200
579
20. 5
1935
3
72. 3
2. 30
0.436
186. 0
2. 1
393
6. 8
4200
554
19. 8
1810
4
100%
13. 5
2. 85
0. 351
34. 2
2. 7
429
11. 5
133. 0
9. 8
U. 0
41. 8
Sacramento
5
River
13. 8
2. 94
0. 341
34. 2
3. 9
455
17. 2
133. 0
13. 2
11. 5
58. 4
Water
6
12. 5
2. 85
0. 351
34. 2
4. 2
417
19. 0
133. 0
8. 2
10. 0
35. 0
7
66. 7%
10. 3
2. 56
0. 391
22. 1
3. 3
548
12. 3
91. 6
9. 7
12. 6
36. 2
Sacramento
8
River
9.4
2. 56
0. 391
22. 1
4. 5
533
16. 8
91. 6
6. 5
U. 0
24. 3
Water
9
8. 8
2. 41
0. 416
22. 1
4. 8
508
16. 6
91. 6
6. 5
10. 3
22. 4
10
33. 3%
4. 4
2. 63
0. 381
9. 7
4. 8
898
18. 5
51. 0
9. 3
10. 6
35. 8
Sacramento
11
River
6. 4
2. 60
0. 385
9. 7
4. 8
1310
18. 3
51. 0
9. 5
15. 4
36. 2
Water
12
4. 2
2. 56
0. 391
9. 7
4. 8
857
17. 9
51. 0
12. 1
10. 8
45. 2
aEstimated from the biomass parameter relationships of Porcella et al. [34],
^ and k^ = 0. 001 day-1.
CK = ^1 ~ ^ where (1 = 1. 85 day"1.
8 (1
-------
cells produced per mg PO4-P removed. This variation was
probably caused by a combination of factors such as the low
precision of the phosphate determination at low concentration,
the poor precision of biomass determination (cell counts) at low
cell levels, and the possibility of particulate matter supplying
phosphorus for cell growth. These values are markedly higher
than those estimated from the phosphate-limiting kinetic study
(Yn for phosphate at a 0 = 2. 6 days was about 350 mg cells
produced per mg PO4-P removed). It is apparent that the data
on phosphorus yield coefficients does not provide any indication
that phosphorus was the growth rate limiting nutrient.
The net yield coefficients based on nitrogen were relatively
constant ranging from 10. 0 to 15.4 with an average of 11.5 mg
cells produced per mg nitrogen removed. These values are
significantly lower than the corresponding values reported by
Porcella ^^al^ [34] (Yn for nitrogen was about 20 ~ 30 mg cells
produced per mg nitrogen removed). It appears that nitrogen was
not the growth rate limiting nutrient. It should be noted that the
possibility existed for the hydrolysis of nitrogenous compounds in
particulate matter present in the samples which would tend to
increase the observed yield coefficients.
Half Saturation Constants, Ks. Using the kinetic data for
Selenastrum capricornutum obtained from the ghosphate-limiting
algal studies (maximum specific growth rate, fx, and decay constant
k^), the half saturation constants for phosphate and nitrogen were
calculated and are presented in Table XXVIII
Nitrogen, Ks. An average half saturation constant of 2140 jig N/?
and 37. 3 fig NIt was obtained for the 30% PAAP control assay
and Sacramento River assays. Unfortunately, there are no reliable
kinetic data for Selenastrum capricornutum with respect to nitrogen.
It is obvious that nitrogen was not a growth-limiting nutrient in the
30% PAAP control. However, based on the Ks data there was a
possibility that nitrogen was the growth rate limiting nutrient in
Sacramento River samples.
Phosphorus, Ks. With respect to phosphorus, average half
saturation constants of 5. 7 p.g PIt and 16. 9 |J.g P It were observed
for the 30% PAAP control and Sacramento River samples,
respectively. Comparing these values with the half saturation
constant obtained from the phosphorus-limiting algal growth kinetic
study (Ks for phosphorus 3. 7 to 5. 8 (ig P/£)» it is apparent that in
the 30% PAAP medium, phosphorus is the primary growth-limiting
nutrient. On the contrary, the high value observed for the Ks of
Sacramento River water indicates that phosphorus definitely is not
the growth rate limiting nutrient.
156
-------
The foreing anlyses indicate the very great utility in the applica-
tion of kinetic constant data in identifying the growth-limiting
nutrient in natural water samples. However to do so, one must
have reliable kinetic data for the nutrient and organism of concern.
Based upon the data obtained from a single series of chemostat
assays of a natural water, the observed data on the half saturation
constant Kg and the yield coefficient, Y, permits a first order
assessment of the likely growth rate limiting nutrients.
Enrichment (Spiking) Studies. To identify the growth-limiting nutrient
by enrichment (spiking) procedures, 3 undiluted Sacramento River
chemostats were operated in accordance with the spiking procedure
outlined in Table XXIX. Daily cell counts and pH observations were
made and these data are also reported in Table XXIX. Figure 51
shows the variation in cell counts as they responded to the different
spiking treatments. In examining the changes of cell counts with time
for the different spiking treatments, in Figure 51 one can conclude
that there were no significant effects on cell counts with spiking
treatment A, B, C, and D. In other words, the cell concentration of
the chemostats were not affected by adding macronutrients (Mg, Ca,
K, S, CI), iron, phosphate, or nitrogen. However, the cell concen-
trations increased about 4 times in about 27 hours after spiking with
macronutrients +Fe + P +N + micronutrients. This indicates that
macronutrients, Fe, P, or N were not the growth-limiting nutrient in
Sacramento River water. However, when micronutrients were added
(plus all others) growth increased markedly, indicating that one of the
micronutrients was limiting the growth rate. Unfortunately, the
specific growth-limiting micronutrient could not be identified because
all micronutrients were lumped together as specified in the standard
PAAP procedure [35]. Unfortunately, the spiking treatment included
all of the macronutrients, including N and P as well as the micronutrients.
It would have been better and the results would have been more
definitive if a spiking treatment of only micronutrients had been used.
Similarly it would be possible to determine which of the micronutrients
was rate-limiting by individual spiking treatments of each micronutrient.
Unfortunately time and funds did not permit such studies.
It should be noted that the batch assay spiking studies did not identify
with certainty the rate-limiting nutrient. However, the chemostat
spiking studies of the same sample indicated clearly the rate-limiting
nutrient. This is additional evidence of the general applicability of the
chemostat in identifying the rate-limiting nutrients, as well as for
determination of the kinetic characteristics of each organism and
nutrient. The application of kinetic information to practical problems
has been described in detail and it appears to be the most promising
method for assessing the general eutrophication problem.
157
-------
TABLE XXIX
CHEMOSTAT ENRICHMENT (SPIKING) PROCEDURE AND OBSERVED
DATA WITH SACRAMENTO RIVER WATER
Day of
Operation
Enrichment
(Spiking) Witha
100% Sacramento River Water
Chemostats No.
Control
Chemostat
1
2
3
pH
Cellb
Count
pH
Cell
Count
pH
Cell
Count
pH
ODd
8
Macronutrientse
8. 2
147
8. 0
150
7. 7
135
8. 3
0. 225
9
Macronutrients
8. 3
176
8. 0
179
7. 9
171
8. 4
0. 230
10
Mac r onutrients
+ Fe
7. 8
189
7. 7
184
7. 9
170
7. 8
0. 200
11
Macronutrients
+ Fe
7. 8
220
8. 2
213
8. 2
190
7. 7
0. 190
12
Macronutrients
+ Fe + P
8. 1
182
8. 3
183
8. 3
168
7. 6
0. 190
13
Macronutrients
+ Fe + P
8. 2
173
8. 3
161
8. 3
150
7. 5
0. 195
14
Macronutrients
¦f Fe + P -f N
8. 3
185
8. 2
176
8. 3
160
7. 8
0. 190
15
Macronutrients
+ Fe + P + N
8. 4
211
8. 3
189
8.4
178
8. 4
0. 190
16
Macronutrients
+ Fe + P + N +
micr onutrients
8. 3
206
8. 3
195
8. 4
170
8. 3
0. 200
17
Macronutrients
+ Fe + P + N +
micr onutrients
8. 4
758
8. 3
800
8. 4
863
8. 4
0. 205
aFinal concentration of spiked nutrients = 30% PAAP medium.
^Cell number x 556/mf (hemocytometer) at time of enrichment.
cControl chemostat operated on 30% PAAP medium; received no enrichment,
^OD - absorbance at 600 m^i.
eMacronutrients including Mg, Ca, K, S, and CI.
fMicronutrients spike identical in composition and concentration to that of 30% PAAP medium.
158
-------
<0
in „„„
m 800
a*
° 600
400
200 —
PRESUMED STEADY STATE SAMPLING DAY,
SPIKING NUTRIENTS - MACRONUTRIENTS0
SPIKING NUTRIENTS = MACRONUTRIENTS + Fe
SPIKING NUTRIENTS = MACRONUTRIENTS + Fe + P
SPIKING NUTRIENTS: MACRONUTRIENTS + Fe +P+N
SPIKING NUTRIENTS = MACRONUTRIENTS + Fe + P+ N
+ MICRONUTRIENTS b
MACRONUTRIENTS INCLUDE Na, K , Mg , CI", SO^,Co ,C0j ft HCO,
MICRONUTRIENTS INCLUDE B , Mn, Z n , Co , Cu 6 Mo
CHEMOSTAT NO. 4
CHEMOSTAT NO. 5
~ CHEMOSTAT NO. 6
I
Hi
o
1-6
1-8 1-9 1-10 HI 1-12 1-13 1-14
DATE (month-day-1971)
1-15 1-16 1-17 1-18
FIGURE 51. SACRAMENTO RIVER CHEMOSTAT CELL
CONCENTRATION, X1( VARIATION WITH
DIFFERENT SPIKING TREATMENTS
159
-------
SECTION VII
COMPARISON OF BATCH AND CHEMOSTAT ALGAL ASSAYS
To define the role of each of the batch and chemostat assays in eutro-
phication analyses it is necessary to compare the assays on the basis
of the defined requirements of algal assays.
GROWTH RESPONSES
Proportionality of Cell Concentration to
Concentration of Growth-Limiting Nutrient
Synthetic Media. The cell concentration parameters X for the batch
assay and Xj for the chemostat assay were found to be significantly
related (p = 0. 99) to the concentration of the growth-limiting nutrient.
However, it was also shown that in batch assays the yield coefficient
(Y) for nitrogen and phosphorus were affected by the initial N/P ratios
in the cultures and in chemostats Y was not constant but a function of
the residence time. Therefore, the linearity of cell concentration
responses is affected by culture conditions and these conditions should
not be ignored.
Natural Water Samples. The cell concentration responses X and Xj
were linearly related (p = 0. 99) with the dilution of the sample, i. e. ,
the more concentrated the sample the greater the cell density. In both
methods this linearity can be used to detect the presence of toxicants
if growth rate is the parameter used; however, in the batch assay small
reductions in the growth rate are not detected while such effects can be
observed easily with chemostat assays. The chemostat is much more
sensitive than the batch assay for the detection of toxic effects.
In general, the basic requirement that algal growth and the resulting
cell concentration should be proportional to the concentration of the
growth-limiting nutrient is fulfilled with both the batch and chemostat
assay method. However, this has been shown to hold only for samples
or media with relatively high concentrations of nutrients (P ^ 30 jig /I
and NH3 -N + N03 -N > 125 \xg/i).
Relationship of Growth Rate With Concentration of Growth-Limiting
Nutrient. The growth rate parameters, |x, for the batch assay and ^
for the chemostat assay were both relatedto the concentration of the
growth-limiting nutrient with the first order—zero order type of relation-
ship (i.e., Michaelis-Menten, Monod equation). However, the
161
-------
relationship only can be estimated crudely in batch assays apparently
due to the transient character of the cell concentration and nutrient
concentrations, whereas the chemostat assays with constant cell con-
centrations and nutrient concentrations permit rather precise deter-
mination of the above relationship. The chemostat method is far
superior to the batch method in determining growth rate—growth limiting
nutrient relationships.
Precision of Growth Responses. The algal growth responses X and (I,
of the batch assay can be determined with reasonable accuracy
(coefficient of variation approximately 15%). However, the chemostat
growth response Xi can be determined with a higher precision (coeffi-
cient of variation 10%). The precision of the chemostat assay appears
to be significantly higher than that of batch assays.
IDENTIFICATION OF THE GROWTH-LIMITING NUTRIENT
Yield Coefficients
In batch cultures the use of yield coefficients (Y) proved to be highly
variable and of virtually no value in determining which nutrient limited
growth. In chemostats Yp was less variable than with batch assays
but it was found to be a function of the residence time (growth rate)
and thus was limited in its use in identifying the growth-limiting
nutrients. However, once the relationship between each nutrient yield
coefficient and residence time is determined (as was done for P in
this study) the use of yield coefficients in chemostat studies may be of
substantial value.
Enrichment (Spiking) Techniques
The value of spiking techniques in batch cultures to identify the growth-
limiting nutrient was variable because the X response was not suffi-
ciently sensitive to identify clearly significant differences in spiking.
Only the response was able to demonstrate clearly that a micro-
nutrient was limiting growth in a specific water, jl^ appears to be
the most sensitive growth parameter for identifying the growth-limiting
nutrient in batch assays by spiking treatments. Moreover, the use
of permits direct interpretations or predictions based upon con-
ceptual eutrophication models such as presented in Figure 1 whereas
the relationship between X and the conceptual model is not at all defined,
that is no theory or empirical knowledge exists on the relation between
X and the cell concentration in natural waters.
Spiking treatments with the chemostats (at a constant residence time)
clearly identified the growth-limiting nutrient in a natural water sample
162
-------
(Sacramento River), Addition of other nutrients caused no significant
increase in Xj , but addition of the growth-limiting nutrient (micro-
nutrients) caused an immediate response in the cell concentration, Xj ,
indicating that growth was limited specifically by a mlcronutrient.
Unlike the batch assays, the chemostat results identified unequivocally
which nutrient limited growth. The chemostat assay appears to be
much more sensitive than the batch assay to spiking treatments in the
identification of the growth-limiting nutrients,
Half Saturation Constants
The batch assay method is not adequate for the accurate determination
of kinetic constants especially the half saturation constants (K )• This
is a severe limitation because of both the significance and role of K
in identifying the growth-limiting nutrients. The chemostat is ideally
suited for determination of Kg values for specific organisms and nutrients.
Comparison of K values obtained with samples for presumed rate-
limiting nutrients and with the true K values determined for the
organism and the actual rate-limiting nutrient permits identification
of the growth-limiting nutrient. This was demonstrated for both the
PAAP medium as well as natural water samples. The ability to use
chemostat-determined K values to identify growth-limiting nutrients
indicates the versatility of the chemostat assay in eutrophication
analyses.
RATIONAL FRAMEWORK FOR THE APPLICATION
OF BIOASSAY RESULTS
The rational application of bioassay results to solve practical eutro-
phication problems requires information on kinetic models to describe
the behavior of natural algal populations, chemical-biological inter-
actions, physical-biological interactions, etc. The most urgently
needed information is accurate kinetic descriptions for the organisms
of concern, and especially for the test alga and the presumed growth
rate limiting nutrients. Information on the growth kinetics of the test
algal species is required to interpret the growth responses obtained by
field sample assays. Knowledge of the growth kinetics of other (nuisance)
algae is needed to extrapolate the bioassay findings to the natural
environment.
The use of the batch culture method to determine the algal growth
kinetics using various models also was found previously to be unsatis-
factory [34]. On the other hand the use of chemostats to determine
the kinetics of growth of Selenastrum capricornutum was found to be
extremely satisfactory (high level of statistical significance) and the
kinetic constants such as pL, Kg, and k^ were estimated accurately and
the dependence of Yp on residence time was evaluated,
163
-------
OPERATIONAL CONSIDERATIONS
Equipment
The equipment needs for chemostat assays are greater than for batch
assays. However, one chemostat unit Is much more versatile than
one batch culture flask, therefore the real difference between the two
methods is probably not significant.
Simplicity
The batch assay is simpler to perform than the chemostat assay.
However, this does not indicate that the needs for operating chemostat
assays require the use of specialized personnel but rather that some
familiarity with the assay method must be developed for satisfactory
operation.
Operational Cost
Although specific cost figures have not been prepared there is probably
very little difference in the operational cost for the two assay methods.
However, chemostat assays require larger volumes of samples than
batch as says.
Time Requirements
The average duration of a PAAP standard batch assay is from 14 to 21
days. To conduct a chemostat assay at a single residence time about
14 days is needed; however, during this period identification of the
growth-limiting nutrient can. be made. With batch cultures the identifica-
tion of the growth-limiting nutrient needs either a separate experiment
for spiking or many different flasks at the beginning; whereas with
chemostats the same units are used for the assay and for identifying
the growth-limiting nutrient.
THE ROLE OF BATCH AND CHEMOSTAT ASSAYS
IN EUTROPHICATION ANALYSIS
Batch Assays
The linearity of cell concentration response with respect to the con-
centration of the growth-limiting nutrient, the reasonably adequate
precision (coefficient of variation approximately 15%) of estimating
biomass responses, the ease of operation, the low equipment cost
164
-------
requirements, etc. , Indicate that batch algal assays should have some
application in eutrophication analysis. However, the inferiority of
the batch assay as compared to the chemostat assay suggests that the
batch assay should be used primarily as a screening device and possibly
as a routine assay for monitoring purposes.
With batch assays both parameters, pi, and X, should be determined.
When spiking treatments are used to aetermine the growth-limiting
nutrient in batch assays, both parameters should be considered as it
appears that fx^ is the most sensitive parameter. Batch assays of rich
natural waters may be subject to growth limitations by carbon and
adequate facilities for ensuring abundant carbon supply should be used
when carbon is not the nutrient of concern. By taking these precautions
the batch culture should be a satisfactory method for screening investi-
gations of natural water samples. However, it must be realized that
the development of a rational framework for the application of batch
assay data, even for initial screening methods is a major factor to be
considered in the selection of the assay method.
Chemostat Assays
Chemostat assays have a dual role in eutrophication analyses [34].
First, they have been found to be extremely suitable both for determin-
ing the algal response of a water sample and for identifying the growth-
limiting nutrient. Identification of the growth-limiting nutrient is
possible by the use of enrichment techniques and/or the determination
of Ks values. Second, the chemostat assay is unquestionably the
method of choice for determining kinetic growth constants for the test
algal s pecies as well as for important (nuisance) algae. Thus, the
chemostat assay provides not only the basic assay growth response
results but also provides a conceptual basis for interpreting these in
terms of the practical eutrophication problem. Kinetic data on the
problem of "nuisance" organism and the rate-limiting nutrient is the
ultimate objective to be reached for the rational analysis and inter-
pretation of control measures for the practical eutrophication problem.
The chemostat assay method is recommended as the preferred method
for laboratory assays and evaluation of the algal growth rate—limiting
nutrient relationship. It is urgent that studies be undertaken to
determine the kinetic characteristics of the test alga Selenastrum
capricornutum with respect to all presumed significant nutrients (this
study provided the first order estimate only for phosphorus). Similarly,
kinetic descriptions for the organisms associated with algal blooms
and the presumed significant nutrients also need to be developed. It
is only with such data that a rational basis for eutrophication control
measures can be developed.
165
-------
SECTION VIII
ACKNOWLEDGMENTS
This report summarizes work accomplished by the University of
California, Berkeley Research Team as a participant in the multi-
laboratory Provisional Algal Assay Procedures Research Program
initiated by the Joint Industry-Government Task Force on Eutrophica-
tion and sponsored by the Environmental Protection Agency.
The cooperation with other laboratories participating in this program
is sincerely appreciated. The sources of results of other participating
laboratories referred to in this report, are gratefully acknowledged.
This research was supported in major part by Research Grant 16010
DQB of the Environmental Protection Agency, U. S. Department of
the Interior. The help provided by Dr. A. F. Bartsch and Mr. Thomas
E. Maloney, the Grant Project Officer, is acknowledged with sincere
thanks,
167
-------
SECTION IX
REFERENCES
1. Hasler, A. D. , "Eutrophication of Lakes by Domestic Sewage, "
Ecology 28, p. 383 (1947).
2. Sawyer, C. N. , "Fertilization of Lakes by Agricultural and
Urban Drainage, " J. New England Water Works Assoc. , 61, p.
109 (1947).
3. Edmundson, W. T. , "Lake Eutrophication and Water Quality
Management: The Lake Washington Case, " Water Quality Control,
Univ. of Washington Press, Seattle, Washington, p. 139 (1968).
4. Thomas, E. A. , "Der Zurichsee, sein Wasser und sein Boden, "
Jahrb. vom Zti'richsee, 17, p. 173 (1957).
5. Mackenthum, K. M. , Keup, L. E. , and Stewart, R. K. , "Nutrients
and Algae in Lake Sebasticook, Maine, " J. Water Pollut. Control
Fed. Research Suppl. , 40, R72 (1968).
6. "Nutrient-Associated Problems in Water Quality and Treatment, "
Task Group 2610-P Report, J. Am. Water Works Assoc. , 58,
p. 1337 (1966).
7. Stewart, K. M. , and Rohlich, G. A. , Eutrophication — A Review,
A Report to the California Water Quality Control Board,
Publication No. 34, p. 188 (1967).
8. National Academy of Science, Eutrophication: Causes, Con-
sequences, Correctives, Proceedings of a Symposium, National
Academy of Sciences, Washington, D. C. , p. 669 (1969).
9. Upsalla Symposium, Eutrophication in Large Lakes and Impound-
ments, O. E. C. D. Report No. 26, p. 560 (1970).
10. Mulligan, H. F. , "Management of Aquatic Vascular Plants and
Algae, " in Eutrophication: Causes, Consequences, Correctives,
National Academy of Sciences, Washington, D. C. , p. 464 (1969).
11. Knight, A. , and Ball, R. C. , "Some Estimates of Primary
Production Rates in Michigan Ponds, " Michigan Academy of
Science, Arts and Letters, 47, p. 219 (1962).
12. Kuentzel, L. E. , "Bacteria, Carbon Dioxide, and Algal Blooms, "
J. Water Pollut. Control Fed. , 41, p. 1737 (1969).
169
-------
13. Legge, R. F. , and Dingledein, D. , "We Hung Phosphates Without
A Fair Trial, " Can. Res. Develop. , p. 19 (March 1970).
14. "Detergent Phosphates Under Fire," Environmental Sci. Tech. , 4,
p. 101 (1970).
15. Bowen, D. H. M. , "The Great Phosphorus Controversy, "
Environmental Sci. Tech. , 4, p. 725 (1970).
16. Sawyer, C. N. , "Discuss ion — Bacter ia, Carbon Dioxide, and
Algal Blooms, " J. Water Pollut. Control Fed. , 42, p. 677 (1970).
17. Abelson, P., "Excessive Emotion About Detergents, " Science,
169, p. 1033 (1970).
18. Kerr, P. C. , Paris, D. F. , and Brochway, D. L. , "The
Interrelation of Carbon and Phosphorus in Regulating Heterotrophic
and Autotrophic Populations in Aquatic Ecosystems, " Water
Pollut. Control Res. Ser. 16050 FGS 07/70, U. S. Environmental
Protection Agency, (1970).
19. Shapiro, J. , "A Statement on Phosphorus, " J. Water Pollut.
Control Fed. , 42, p. 772 (1970).
20. Kuentzel, L. E. , "Correction, 11 J. Water Pollut. Control Fed. ,
42, p. 353 (1970).
21. Dawson, W. L. , Phosphates in Detergents and the Eutrophication
of America's Waters, Twenty -Third Report by the Committee on
Government Operations, 91st Congress, 2nd Session, House
Report No. 91-1004, U.S. Government Printing Office, Washington,
D. C. (1970).
22. King, D. L. , "The Role of Carbon in Eutrophication, " J. Water
Pollut. Control Fed. , 42, p. 2035 (1971).
23. Likens, G. E. , "The Limiting Nutrient Controversy, " Symposium
on Nutrients and Eutrophication, Am. Soc. Limnol. and Oceanog. ,
Michigan State University, East Lansing, Michigan (1971).
24. Ryther, J. H. , and Dunstan, W. M. , "Nitrogen, Phosphorus, and
Eutrophication in the Coastal Marine Environment, " Science, 171,
p. 1008 (1971).
25. Monod, J. , "The Growth of Bacterial Cultures, " Ann. Rev.
Microbiol. , 3, p. 371 (1949).
26. Caperon, J. W. , The Dynamics of Nitrate Limited Growth of
Isochrysis galbana Populations, Ph. D. Thesis, University of
California, San Diego (1965).
170
-------
2 7. Dugdale, R. C. , "Nutrient Limitation in the Sea: Dynamics,
Identification and Significance, " Limnol. and Oceanog. , 12,
p. 685 (1967).
28. Eppley, R. W. , Rogers, J. N. , and McCarthy, J. J. , "Half
Saturation Constants for Uptake of Nitrate and Ammonium by
Marine Phytoplankton, " Limnol. and Oceanog. , 14, p. 912 (1969).
29. Smith, F. E. , "Effects of Enrichment in Mathematical Models, "
in Eutrophication: Causes, Consequences, Correctives, National
Academy of Sciences, Washington, D. C. , p. 631 (1969).
30. McGauhey, P. H. , Rohlich, G. A. , Pearson, E. A. , Tunzi, M. ,
Aleti, A., and Middlebrooks, E. J. , First Progress Report,
Eutrophication of Surface Waters — Lake Tahoe Bioassay of
Nutrient Sources, Lake Tahoe Area Council, South Lake Tahoe,
California, p. 178 (1968).
31. McGauhey, P. H. , Rohlich, G. A. , Pearson, E. A. , Middlebrooks,
E. J. , Porcella, D. B. , Aleti, A. , and Tunzi, M. , Eutrophication
of Surface Waters — Lake Tahoe: Laboratory and Pilot Pond
Studies, Lake Tahoe Area Council, South Lake Tahoe, California
p. 180 (1969).
32. McGauhey, P. H. , Pearson, E. A. , Rohlich, G. A. , Porcella, D. B.
Dugan, G. L. , and Middlebrooks, E.J. , Eutrophication of Surface
Waters — Lake Tahoe (Indian Creek Reservoir), Lake Tahoe Area
Council, South Lake Tahoe, California, p. 141 (1970).
33. McGauhey, P. H. , Pearson, E. A. , Rohlich, G. A. , Dugan, G. L. ,
Porcella, D. B. , and Middlebrooks, E.J. , Eutrophication of
Surface Waters — Lake Tahoe: Pilot Pond and Field Studies,
Lake Tahoe Area Council, South Lake Tahoe, California, p. 109i
(1970).
34. Porcella, D. B. , Grau, P. , Huang, C. H. , Radimsky, J. ,
Toerien, D. F. , and Pearson, E. A. , Provisional Algal Assay
Procedures: First Annual Report, SERL Report No. 70-8,
University of California, Berkeley, California, (1970).
35. Provisional Algal Assay Procedures, Joint Industry — Government
Task Force on Eutrophication, P.O. Box 3011, Grand Central
Station, New York, p. 62 (1969).
36. Fitzgerald, G. P. , "Field and Laboratory Evaluations of Bioassays
for Nitrogen and Phosphorus with Algae and Aquatic Weeds, "
Limnol and Oceanog. , 14, p. 206 (1969).
37. Eppley, R. W. , and Coatsworth, J. L. , "Uptake of Nitrate and
Nitrite by Ditylum bughtwellii — Kinetics and Mechanisms, "
J. Phycol. , 4, p. 151 (1968).
171
-------
38. Holm-Hansen, W. , Booth, C. R. , "The Measurement of Adenosine
Triphosphate in the Ocean and its Ecological Significance, " Limnol.
and Oceanog. , 11, p. 510 (1966).
39- Strickland, J. D. H. , and Parsons, T. R. , A Manual of Sea Water
Analysis, Bulletin No. 125, Fisheries Research Board of Canada,
Ottawa (1968).
40. Lorenzen, C. J. , "A Method for the Continuous Measurement of
m vivo Chlorophyll Concentration, " Deep Sea Research, 13, p. 22 3
(19W
41. Manny, B. A. , "The Relationship Between Organic Nitrogen and
The Carotenoid to Chlorophyll a Ratio in Five Freshwater Phyto-
plankton Species, " Limnol. and Qceanog. , 14, p. 69 (1969).
42. Gaarder, T. , and Gran, H. H. , "Investigations of the Production
of Plankton in the Oslo Fjord, " Rapp. et Proc. - Verb. , Cons.
Int. Explor. Mer. , 42, p. 1 (1927).
43. Steemann-Nielsen, E. , "The Use of Radio-Active Carbon (C14)
For Measuring Organic Production in the Sea, " J. Cons. Int.
Explor. Mer., 18, p. 1 17 (1952).
44. Goldman, C. R. , "Micronutrient Limiting Factors and Their
Detection in Natural Phytoplankton Populations, " Memorie 1st.
Ital. Idrobiol. , Suppl. , 18, p. 121 (1965).
45. Allen, E. J. , and Nelson, E. W. , "On the Artific ial Culture of
Marine Plankton Organisms, " J. Mar. Biol. Assoc. , 8, p. 421
(1910).
46. Eyster, C. , "Bioassay of Water From a Concentration-Forming
Marl Lake, 11 Limnol. and Qceanog. , 3, p. 455 (1958).
47. Bringmann, G. , and Ktlhn, R. , "Der Algan-Titer als Mas stab
der Eutrophierung von Wasser und Schlamm, " Gesundheitsingenier.
79., p. 50 (1956).
48. Skulberg, O. M. , "Algal Cultures as a Means to Assess the
Fertilizing Influence of Pollution," In: Advances in Water Pollution
Research, Water Pollution Control Federation, Washington, D. C. ,
Vol. 1, p. 113 (1967).
49. Oswald, W. J. , and Golueke, C. G. , "Eutrophication Trends in the
United States. A Problem?" J. Water Pollut. Control Fed. , 38,
p. 964 (1966).
50. Bain, Jr., R. C. , "Algal Growth Assessments by Fluorescence
Techniques, " In: Proceedings of the Eutrophication-Biostimulation
Assessment Workshop, Ed. E.J. Middlebrooks, et al. , p. 37 (1969).
172
-------
51. Goldman, C. R. , Tunzi, M. G. , and Armstrong, R. , Ml4C Uptake
as a Sensitive Measure of the Growth of Algal Cultures, " In:
Proceedings of the Eutrophication-Biostimulation Assessment
Workshop, Ed" E.J. Middlebrooks, et al. , p. 158 (1969).
52. Pearson, E. A. , The Case for Continuous Flow (Chemostat)
Kinetic Descriptions of Plankton-Nutrient Growth Relationships
in Eutrophication Analyses, Prepared for Joint Industry Govern-
ment Committee Meeting on Algae Growth Potential, Chicago,
Illinois (6-8 March 1968).
53. Pearson, E. A. , Middlebrooks, E.J. Tunzi, M. , Aleti, A.,
McGauhey, P. H. , and Rohlich, G. A. , "Kinetic Assessment of
Algal Growth, " Proceedings of the Eutrophication-Biostimulation
Assessment Workshop, Ed. E. J. Middlebrooks, et al. , p. 56
(1969).
54. Borchardt, J. A. , "Biological Extraction of Nutrients, " J. Water
Pollut. Control Fed. , 40, p. 1739 (1968).
55. Scherfig, J. , and Dixon, P. S. , Results presented at PAAP
Evaluation Meeting, Madison, Wisconsin, (October 1970).
56. Miller, W. E. , Results presented at PAAP Evaluation Meeting,
Madison, Wisconsin, (October 1970).
57. Lindemann, E. , and Vollp, G. P. , Eutrophication: Progress
Report, Report No. PCR 762, FMC Corporation, Princeton (1970).
58. Haviland, J. R. , Wilkes, F. G. , Sless, J. B. , Marshall, G. H. ,
and Weiss, C. M. , Provisional Algal Assay Procedures: Interim
Report No. 2, Department of Environmental Sciences and Engineering,
University of North Carolina, Chapel Hill (1970).
59. Williams, J. M. , andYeisley, W. G. , Provisional Algal Assay
Procedure: Eutrophication Status Report, Lever Brothers,
Edgewater, New Jersey (1970).
60. Payne, A. G. , Results presented at PAAP Evaluation Meeting,
Corvallis, Oregon (May 1970).
61. Sless, J. B. , Haviland, J. R. , Wilkes, F. G. , Marshall, G. H. ,
and Weiss, C. M. , Provisional Algal Assay Procedures: Interim
Report, Department of Environmental Sciences and Engineering,
University of North Carolina, Chapel Hill (1969).
62. Haviland, J. R. , Wilkes, F. G. , Marshall, G. H. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 3, Depart-
ment of Environmental Sciences and Engineering, University of
North Carolina, Chapel Hill (1970).
173
-------
63. HavLland, J. R. , Wilkes, F. G. , Marshall, G. H. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 4,
Department of Environmental Sciences and Engineering, University
of North Carolina, Chapel Hill (1970).
64. Wilkes, F. G. , Haviland, J. R. , Marshall, G. H. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 5,
Department of Environmental Sciences and Engineering, University
of North Carolina, Chapel Hill (1970).
65. Haviland, J. R. , Wilkes, F. G. , Marshall, G. H. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 6,
Department of Environmental Sciences and Engineering, University
of North Carolina, Chapel Hill (1970).
66. Haviland, J. R. , Marshall, G. H. , and Weiss, C. M. , Provisional
Algal Assay Procedure: Interim Report No. 7, Department of
Environmental Sciences and Engineering, University of North
Carolina, Chapel Hill (1970).
67. Haviland, J. R. , Wilkes, F. G. , Marshall, G. H. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 8,
Department of Environmental Sciences and Engineering, University
of North Carolina, Chapel Hill (1970).
68. Marshall, G. H. , Haviland, J. R. , Wilkes, F. G. , and Weiss, C. M. ,
Provisional Algal Assay Procedure: Interim Report No. 9.
Department of Environmental Sciences and Engineering, University
of North Carolina, Chapel Hill (1970).
69. Weiss, C. M. , Helms, R. W. , and Marshall, G. H. , The Inter -
laboratory Precision Test: An Eight laboratory Evaluation of
the Provisional Algal Assay Procedure Bottle Test, Department
of Environmental Sciences and Engineering, University of North
Carolina, Chapel Hill (19711.
70. Payne, A. G. , Results presented at the PAAP Evaluation Meeting,
Madison, Wisconsin, (October 1970).
71. Gerhold, R. M. , Effects of Inoculum Age on Growth Rates and
Final Yields in Selenastrum capricornutum Cultures in Full
Strength and Dilute October '69 PAAP Medium, Water Resources
Center, Univers ity of Wisconsin, Madison, W iscons in (1970V
72. Gerhold, R. M. , and Uttormark, P. D. , Preliminary Development
of Flask Assay Procedures, Water Resources Center, University
of Wisconsin, Madison, Wisconsin (1970).
73. Maloney, T. E. , Summary of PAAP Evaluation Meeting, October
8 - 9, 1970, National Eutrophication Research Program, Environ-
mental Protection Agency, Corvallis, Oregon (1970).
174
-------
74. Fitzgerald, G. P. , Communication made at the PAAP Evaluation
Meeting, Madison, Wisconsin, (October 1970).
75. Yeisley, W. G. , Results presented at the PAAP Evaluation Meeting,
Madison, Wisconsin, (October 1970).
76. Scherfig, J. , Results presented at the PAAP Evaluation Meeting,
Corvallis, Oregon, (May 1970).
77. Maloney, T. E. , Algal Assay Procedure: Bottle Test, National
Eutrophication Research Program, Environmental Protection
Agency, Corvallis, Oregon (1971).
78. Toerien, D. F. , Huang, C. H. , Porcella, D. B. , Radimsky, J. ,
and Pearson, E. A. , The Role of Chemostat Assays in Eutrophica-
tion Analysis, Presented at the 21st AIBS Meeting, Bloomington,
Indiana (1970).
79. Fencl, Z. , "A Uniform System of Basic Symbols for Continuous
Cultivation of Microorganisms, " Folia Microbiol. , 8, p. 192
(1963).
80. Fencl, Z. , "Theoretical Analysis of Continuous Culture Systems,"
In: Theoretical and Methodological Basis for Continuous Culture
of Microorganisms, Ed. I. Malek and Z. Fencl, Academic Press,
New York (1966).
81. Michaelis, L. , and Menten, M. L. , "Die Kinetlk der Invertinwirkung, "
Biochem. Z. , 49, p. 333 (1913).
82. Monod, J. , Reserches sur la Croissance des Cultures Bacteriennes,
Paris, Hermann et C^e (1942).
83. Stewart, M. J. , Reaction Kinetics and Operational Parameters of
Continuous Flow Anaerobic Fermentation Processes, SERL
Publication No. 4, IER Series 90, University of California,
Berkeley (195 8).
84. Agardy, F. J. , Cole, R. D. , and Pearson, E. A. , Kinetics and
Activity Parameters of Anaerobic Fermentation Systems, SERL
Report No. 63-2, University of California, Berkeley (1963).
85. Andrews, J. F. , Cole, R. D. , and Pearson, E. A. , Kinetics and
Characteristics of Multistage Methane Fermentations, SERL
Report No. 64-11, University of California, Berkeley (1964).
86. Teissier, G. , "Les Lois Quantitatives de la Croissance, " Ana.
Physiol. Physiocochim. Biol. , 12, p. 527 (1936).
87. Schulze, K. L. , and Lipe, R. S. , "Relationship Between Substrate
Concentration, Growth Rate, and Respiration Rate of Escherichia
coli in Continuous Culture, " Archiv Mikrobiol. , 48, p^ 1 (1964).
175
-------
88. Schulze, K. L. , "The Activated Sludge Process as a Continuous
Flow Culture, " Water Sew. Works, 111, p. 526 (1964).
89. Schulze, K. L. , "The Activated Sludge Process as a Continuous
Flow Culture, " Water Sew. Works, 112, p. 1 1 (1965).
90. Shelef, G. , Oswald, W. J. , and Golueke, C. G. , Kinetics of Algal
Systems in Waste Treatment — Light Intensity and Nitrogen Con-
centration as Growth Limiting Factors, SERL Report No. 68-4,
University of California, Berkeley (1968).
91. Moser, H. , The Dynamics of Bacterial Populations Maintained
in the Chemo"stat, Publication No. 614, Carnegie Institute of
Washington (1958).
92. Contois, D. E. , "Kinetics of Bacterial Growth — Relationship
Between Population Density and Specific Growth Rate of
Continuous Cultures," J. Gen. Microbiol., 21, p. 40 (1959).
93. Gerloff, G. C. , and Skoog, F. , "Nitrogen as a Limiting Factor
for the Growth of Microcystis aeruginosa in Southern Wisconsin
Lakes, " Ecology, 38, p. 556 (1957).
94. Birge, E. A. , and Juday, C. , "The Inland Lakes of Wisconsin —
The Plankton. I. Its Quantity and Chemical Composition, "
Bull. Wis. Geol. Nat. Hist. Surv. , 64, p. 219 (1922).
95. Mackenthum, K. M. , and Ingram, W, M. , Biological Associated
Problems in Fresh Water Environments, USDI, FWPCA (1967).
96. Fuhs, G. W. , "Phosphorus Content and Rate of Growth in the
Diatoms Cyclotella nana and Thalassiosira fluvlalilis, J. Phycol.
5, p. 312 (1969).
97. Rodhe, W. , "Environmental Requirements of Fresh-Water
Plankton Algae, " Symb. bot. Upsal. , 10, p. 1 (1948).
98. Mackereth, F. J. H. , "Phosphorus Utilization by Asterionella
formosa Hass, " J. Exp. Bot., 4, p. 296 (1953).
99. Marr, A. G. , Nilson, E. H. , and Clark, D. J. , "The Maintenance
Requirements of Escherichia coli, " Ann. N. Y. Acad. Sci. , 102,
p. 536 (1963).
100. Hetling, L. J. , Washington, D. R. , and Rao, S. S. , Kinetics of the
Steady State Bacterial Culture. II. Variations in the Rate of
Synthesis, Rennselaer Polytech. Inst. Troy, New York (1964).
101. Mor, J. R. , and Fiechter, A., "Continuous Cultivation of
Sac char omyc e s cerevls lae. I. Growth on Ethanol under Steady-
State Conditions, " Bjotech. Bioeng. , 10, p. 159(1968).
176
-------
102. Mor, J. R. , and Fiechter, A. , "Continuous Cultivation of
Saccharomyces cerevis iae. II. Growth on Ethanol under Trans ient-
State Conditions, " Biotech. Bioeng. , 10, p. 787 (1968).
103. Zabat, M. , Kinetics of Phosphorus Removal by Algae, Ph. D.
Thesis, University of California, Berkeley (1971).
104. Goldberg, E. D. , Walker, T. J. , and Whisenand, E. , "Phosphate
Utilization by Diatoms, " Biol. Bull. 101, p. 274 (1951).
105. Strickland, J. D. H. , and Parsons, T. R. , A Manual of Sea Water
Analysis, Bulletin No. 125, Fisheries Research Board of Canada,
Ottowa (1968).
106. Standard Methods for the Examination of Water and Waste Water,
12th Ed. , American Public Health Association, New York (1965).
107. Solorzano, L. , "Determination of Ammonia in Natural Waters
by the Phenolhypochlorite Method, " Limnol. Oceanog. , 14,
p. 799 (1969).
108. Maloney, T. E. , Summary of PAAP Evaluation Meeting at
Corvallis, May 1970, National Eutrophication Program, Corvallls,
Oregon (1970).
109. Dixon, P. S. , Personal Communication (1970).
110. Water Quality Criteria, Report of the National Advisory
Committee to the Secretary of the Interior, Federal Water
Pollution Control Administration, Washington, D. C. (1968).
111. Holm-Hansen, O. , "Environmental and Nutritional Requirements
for Algae, " In: Proceedings of the Eutrophication-Biostimulation
Assessment Workshop, Ed. , E. J. Middlebrooks, et al. , p. 98
(1969).
177
-------
SECTION X
PUBLICATIONS AND PATENTS
Publications
1. Porcella, D. B. , Grau, P. , Huang, C. H. , RadLmsky, J. ,
Toerien, D. F. , and Pearson, E. A. , "Provisional Algal
Assay Procedures — First Annual Report, " SERL Report
No. 70-8, Sanitary Engineering Research Laboratory,
University of California, Berkeley (1970).
2. Toerien, D. F. , Huang, C. H. , Porcella, D. B. , Radimsky, J. ,
and Pearson, E. A. , "The Role of Chemostat Assays in
Eutrophication Analysis, " Presented at the 21st AIBS Meeting,
Bloomington, Indiana (1970).
Patents
None
179
-------
SECTION XI
GLOSSARY
Nutrient content of individual cell at cell age 0 in
chemostat
Nutrient content of individual cell at infinite cell age in
chemostat
Growth parameter in Contois' function
Coefficient of variation
Flow rate, volume/time
Fraction of excess nutrient (phosphorus) uptake
Generation time, time
A growth contant in Teissier's Exponential Model
Nutrient utilization rate, time"1
Specific organism decay rate, time"1
Nutrient utilization constant in exponential function, time
Nutrient utilization constant in hyperbolic function, time
A growth constant in modified Teissier's Model, mass/
volume
A rate constant in Blackman First Order Function
Half saturation constant, mass/volume
Number of events
Number of generations
Probability of occurrence of the event
Population density
Rate of predation of algal cells
Nutrient removal velocity, time"1
Essential nutrient concentration, mass/volume
Nutrient concentration in chemostat or in the effluent,
mass/volume
Initial nutrient concentration, mass/volume
Suspension rate of benthic algae
Nutrient saturation concentration, mass/volume
Excess nutrient uptake concentration, mass/volume
Nutrient concentration utilized by cells for their growth,
mass/volume
181
-------
t
t
e
V
X
X
o
Xj
X2
x3
x4
A
x
x
]
Y
Y
i
Y
e
e
max
n
max
w
A
A
Net settling rate of algal cells
Resuspension rate of algal cells
Time
Time length of batch growth to reach zero growth rate
Chemostat volume, volume
Organism concentration, mass/volume
Initial concentration of organisms, mass/volume
Organism concentration in chemostat or effluent, mass/
volume
Concentration of higher organisms responsible for algal
predation
Algal cell concentration available for predation by higher
organisms
Concentration of planktonic algae on the bottom of water
body
Concentration of benthic algae
Maximum concentration of organisms, batch, mass/
volume
Maximum steady state cell concentration in chemostat,
mass/ volume
Growth yield coefficient
Net cell yield coefficient
Maximum cell yield coefficient
Hydraulic residence time, time
Mean cell age, time
Residence time at which "washout" occurs, time
Specific growth rate, time-1
Specific growth rate, batch, time-1
Maximum specific growth rate, time"1
Maximum specific growth rate, batch, time-1
182
-------
SECTION XII
APPENDICES
Page No.
A. Air Supply System for Batch and Chemostat
Cultures 184
B. Computer Program G2 CAL NUN 186
C. Batch Culture Assay Results 196
Table C-I: First Interlaboratory Precision
Test 196
Table C-II: Second Interlaboratory Precision
Test 197
Table C-III: Clear Lake Study 198
Table C-IV: Secondary Effluent Enrichment
of Clear Lake Water 199
Table C-V: Clear Lake Enrichment
(Spiking) Tests 200
Table C-VI Sacramento River Study 202
Table C-VII Secondary Effluent Enrichment
of Sacramento River Water .... 203
Table C-VIII Sacramento River Enrichment
(Spiking) Tests 204
D. Chemostat Operation Data 206
183
-------
APPENDIX A
AIR SUPPLY SYSTEM FOR BATCH AND
CHEMOSTAT CULTURES
1. Flow rates required.
a) 0.12 ft3 hr-1 in each batch culture
b) 0.50 ft3 hr"1 in each chemostat culture.
It is exoected that it may be necessary to add carbon dioxide to
the air stream and/or to vary the flow rates to maintain the pH
in the desired range of 7.0 to 8.3.
2. Mean residence times in glass wool filters.
The mean residence times can be calculated as illustrated:
Assume minimum .reasonable decontamination occurs in 0.25 minute.
Then for a 2-inch diameter PVC pipe:
Area, A = 19.6 cm2 = 20 cm2
Total Air Flow Rate, Q = (# batch)(0.12) + (# chemostats)
(0.5 ft3 hr-1
= (25)(0.12) + 16(0.5)
= 11 ft3 hr"1
= 312 i hr"1
= 312 x 103 cm3 hr-1
Velocuy, » - j} .
- 16 x 103 cm hr"1
= 270 cm min"1
For Dlug flow the residence time in a 2-inch diameter PVC pipe,
t = L/v
0.25 - 4
184
-------
L = (0.25)(270) cm
= 68 cm long tube
to provide for variation a 90 cm tube can be used.
3. Filtration, trapping and humidification.
a) Glass wool — particulate filtration (replace every 6 months).
b) Activated carbon - organic trap (replace every 6 months).
c) 0.1 N H2S01+ in distilled water — trap NH3 (replace after 20
Brcent loss of acidity).
d) Glass bead trap directly after acid wash - prevent carryover of
acid.
e) Distilled water trap - humidify (replace after 20 percent loss
of volume).
f) Glass bead trap after water trap - prevent carryover of water
droplets.
g) Mixing flask for air and carbon dioxide after water trap and
glass bead trap.
4. Tubing size and flow measurement.
a) The manifold size specified will provide adequate flows for
tubing sizes <0.5 mm ID. Tygon tubing should be *6 mm (1/4 in.)
ID.
b) At least one flowmeter for the batch manifold and one for the
chemostat manifold must be provided. The individual chemostats
should each have a flowmeter.
185
-------
APPENDIX B
COMPUTER PROGRAM G2 CAL NLIN
IDENTIFICATION
Code Designation:
Title:
Author:
Date:
Speci fi cati on:
Decks Included:
Additional Decks
Needed:
G2 CAL NLIN
Nonlinear Regression
R. M. Baer, Computer Center
University of California, Berkeley
June 1967
CDC 6400 SCOPE 2.0 Operating System
TESTL (main program)
INDATA
SETUP
GAUSS
DMPMAT
DERIV
BTEST
FINALE
LINEOF
YCOMP (written and supplied by the user)
PURPOSE
To find the values of the parameters Blt B2 which minimize
the sum of squares
S = l[Z| • G(t Z3»•••J Bj > B2!•••)]
where the summation is over a collection of "data points," each point
consisting of a set of "observed" values for the variables (zj,...,zm),
*0r the relative sum of sguares (either with respect to the observed
values Zx or the calculated values G(...) ).
186
-------
m > 2, where the fitting parameters Bls and the observables
2X, z2>... satisfy an equation
z! = G(z2,...; Bi.Ba,...).
Alternately the program may be used to find a "best" solution to a set
of nonlinear equations for which the number of equations is greater
than or equal to the number of unknowns.
METHOD
The procedure which is used here replaces the function G by its
first order Taylor expansion (in the B,-), solves for the minimum of S
(which is being approximated by a quadratic) by solving the set of
linear equations which express the fact that the approximation of S
should have zero gradient. The values of the parameters B,- are changed
accordingly; and then the whole procedure is iterated until the correc-
tions for the B,- become negligible or the number of iterations exceeds
a limit.
References
1. Collatz, L., Numerische und graphische Methoden. Handbuch d.
Physik, Band II, Berlin (1955).
2. Hartley, H. 0., The Modified Gauss-Newton Method for the Fitting of
Non-linear Regression Functions by Least Squares. Technometrics
3,269-280 (1961).
3. Baer, R. M., Nonlinear Regression and the Solution of Simultaneous
Equations. Comm. ACM 5, No. 7 (1962).
USAGE
I. The Complete NLIN Program
The embedding routine (main routine) of this program has as its
purpose 1) introducing and checking options and data, and 2) calling
the two major subroutines, GAUSS (along with SETUP) and FINALE.
GAUSS is the subroutine which seeks the minimizing values of the
parameters, and FINALE is the subroutine which produces a tabulation of
the results of fitting for current values of the parameters Bj.
187
-------
Either of these subroutines may be used independently for specific
needs: in solving certain automatic control problems, programs may use
only GAUSS and SETUP since a tabulation of the goodness of fit at each
fitting would result in too much output and would slow down the control
nrogram; on the other hand, some users want to see merely a tabulation
of the fitting of data by an equation which is nonlinear in the para-
meters and so use only FINALE - the parameters in their case have fixed
values (e.g., are certain natural constants) and what is of interest is
how good the data are. The usage of these subroutines is covered in
later sections. The following two sections cover the usage of the complete
library program.
A. Data Input to the Embedding Program (Library Version)
Each problem to be processed by NLIN requires a data deck,
described below. Several problem decks may be stacked. Processing
will be terminated by a control card carrying a zero. The data
deck consists of
i) a control card - See B.l
ii) card(s) with starting guesses for the parameters
Bj, B2,... - See B.2
iii) cards with the values of the "data point"
observables Z^N), Z2 (N),... n=l, 2, ....
NPTS. (The data point cards may be omitted.
See B.l, B.3)
B. Data Entries
1. Each problem requires its own control card, from which
the program reads ten control values. These ten (integer
type) quantities are read according to FORMAT (1016). The
program identifiers for these quantities are (CNTROL. (I),
1=1, 10). The values must be set in accordance with the
fol1owi ng:
CNTROL (1)
CNTROL (2)
(on option)
(on option)
is the number of parameters B^ with respect
to which the sum of squares is to be mini-
mized. If this variable is given a positive
value then the program expects to read a data
card carrying guesses at the values of the B-j.
If a minus sign is prefixed to the value of
CNTROL(1) then the program skips the command
which reads in values for the B-j (and so
there should be no cards carrying values for
the Bj).
is the number of data points, Z, which are to
be used in the minimization. If CNTR0L(2) is
188
-------
CNTROL (3)
CNTROL (4)
CNTROL (5)
CNTROL (6)
positive, the program expects to read data
cards carrying the values of the Z-j. If
CNTROL(2) has a minus sign prefixed to its
value, then the program skips the commands
which read values for the data points. For
example, a data point might consist of
(Z1,Z2,Z3); and there might be 4 data points
in the particular problem. Then the value of
CNTR0L(2) would be 4, unless the same data
points are to be used as may have already been
read-in from a preceding problem - in which
case CNTROL(2) may be given the value -4 and
duplicating the data cards carrying values for
the data points may be skipped.
is the number of Z^ per data
Example each data point is a
so that CNTROL(3) is 3.
point; in the
triplet (ZlsZ2,Z3)
is an upper bound on the number of iterations.
(A conservative value is 5 + 2k where k is the
number of Bj).
determines the printout of intermediate
iterations as follows: Let N = J + 2K + 4L,
where J, K, and L can be either 0 or 1. J
refers to the printout of input data, K to
the printing of consecutive iterations (sum
of squares and parameters B), and L to the
current matrices for the normal equations.
A value of £ indicates that no printing is
desired; a value of 1_ indicates that the inter-
mediate printing is to be done. Set CNTROL(5)
= N.
determines whether the sum of squares, or a
relative sum of squares is to be minimized.
The value 0 results in minimization of the
ordinary sum of squares; the value 1 results
in the minimization of the relative sum of
squares with respect to the observations; -1,
results in the minimization of the relative sum
of squares with respect to the calculated
functi on.
CNTROL (7) is set by the program and equals the current
iteration number.
CNTROL (8) is set by the program and indicates the nature
of termination.
189
-------
*CNTROL (9) is at the user's disposal and may be used to
indicate the computation (i.e., the fitting
function) which is to be used.
CNTROL (10) is at the user's disposal. A value of £ permits
free correction of the parameters at each
iteration. A value of permits the parameters
to be corrected by at most ten percent per
iterati on.
2. Parameter Entries are (in format 8F10.4):
On the control card the variable NB represents the number of
parameters to be fitted. The card(s) carrying starting
guesses for these parameters immediately follow the control
card.
3. Data Point Entries for each data point, n, are (in format
8F10.4):
Zt(N) Z2(N) ... Zm(N)
where Zx(N) is the entry for the dependent variable and
where m is the number of entries for each data point (with
a maximum m = 8 corresponding to the Dimention Z(8,200) ).
The routines will compute on the basis of a matrix of all
the data, viz.,
ZjO), z2(i),...,zm(i)
for the first data point, and
Z2 (2), Zz(2) Zm(2)
for the second data point, etc., until finally
ZjfNPTS), Z2(NPTS),...,Zm(NPTS)
for the last data point.
On the control card the variable NPTS controls the number
of data points to be read and/or fitted. A maximum of 200
points is permissible with the library version of NLIN.
*A negative value of CNTROL(9) results in an output format suitable for
teletype when Common File version of NLIN is used.
190
-------
II. Function YCOMP
The user must supply a function routine representing the nonlinear
equation(s)
Zj = G(z2 ,...; BjjBg,...)
for his particular problem.
The function must be a (FORTRAN IV) function subprogram named YCOMP with
the following basic form:
FUNCTION YCOMP (N,B,Z)
DIMENSION B(24), Z(8,200)
YCOMP = The expression for G(z2,...; Bl9B2,...)
RETURN
END
YCOMP is computed for the n-th data Doint: thus the fixed point argu-
ment N in the calling sequence. All references to the zq-(N) must be
of the form: Z{2,N), Z(3,N), etc. All references to the Bj must be
of the form: B(l), B(2), etc. The array dimensions for z-jXN) must
agree with those of the embedding program. Those given are for the
library version.
III. Example
Suppose that four experiments have been performed in which flux,
denoted F, is measured as a function of two field strengths, denoted E
and H. A tabulation of the data gives
F E H
1.6667 1.0000 2.0000
1.5000 2.0000 3.0000
1.4000 3.0000 4.0000
1.3333 4.0000 5.0000
It is known that the relation among the variables is given by
Bi + B3H
F = B2 + E
There is some uncertainty as to the values of the parameters Bls B2, and
B3 in this expression but it is believed that all three have values not
too far from 1.5. If GAUSS, together with its accompanying subprograms
and embedding program, were to be used to find least-squares values for
191
-------
Bj, the usual deck would probably consist of a small FORTRAN source deck
(for YCOMP), followed by the binary decks for the embedding program and
GAUSS and FINALE (and DERIV), the binary deck for F4 CAL LINEQS, and
finally (control and) data cards. At present the binary deck for NLIN
contains LINEQF, so that the user need only add his YCOMP, control, and
data cards.
Thus, there are three parameters B, four data points, three variables
(F, E, and H). Suppose we want a limit of 40 iterations, print-out of
the input data and sum of squares and parameters and current matrices of
normal equations, at each iteration (so that CNTR0L(5) = 7), and minimi-
zation of sum of squares of relative errors.
YCOMP and the control and data cards would appear as follows:
FUNCTION YC0MP(N,B,Z)
DIMENSION B(24), Z(8,200)
YCOMP = (B(1) + B(3) * Z(3,N))/(B(2) + Z(2,N))
RETURN
END
7-8-9 card to introduce data
3 4
3.40
7 CONTROL
1.5
1.5
1.5 B(J) starting guesses
1.6667
1.0
2.0 Z(I,1) observations for point
1
1.5
2.0
3.0 Z(I,2) observations for point
2
1.4
3.0
4.0 Z(I»3) observations for point
3
1.3333
4.0
5.0 Z(I,4) observations for point
4
0 CONTROL STOP
A comparison between NLIN's operation when supplied exact data and when
given less exact data is available in the writeup for 62 BC NLIN. There
is also a comparison of performance when a large number of parameters
to be fitted is used instead of a small number.
IV. Sources of Error
A. Overflows
Overflow from products and sums may be caused by exceedingly
bad guess for the B(J).
The use of very large parameters or data may cause overflows
also. Scaling the input data so that the Z2(N) are close to
unit may help.
B. Singularity of normal equations is usually due to a miscoded
function (such as the inclusion of Some Bj value which is not
a parameter).
192
-------
C. The final values of B(J) are the best found by GAUSS but not
necessarily the solution to the problem.
V. Subroutine GAUSS
The duty of subroutine GAUSS is to find values for the parameters
Bj which minimize the sum of squares of differences between the fitting
functions and the dependent observable. GAUSS may find the minimizing
values and return, or may find better values than the starting guesses
but be forced, by the limit on the number of iterations, to return
before the minimizing values have been discovered, or may be stymied by
the occurrence of a set of normal equations which cannot be solved. In
any case, upon return of control from GAUSS to the calling routine the
set of values returned for the parameters Bj are always those giving
whatever minimum sum of squares has been discovered.
GAUSS is invoked by the call
CALL GAUSS (CNTROL, B,Z, MDIM, NDIM, BDIM, A)
where
MDIM, NDIM, BDIM, and CNTROL are fixed-point variables and B,
Z, and A are floating point variables.
Subroutine SETUP is invoked by the call
CALL SETUP (CNTROL,B,Z,SUMSQ,MDIM,NDIM,BDIM)
with the variables essentially the same as above. Subroutines
GAUSS and SETUP set up the normal equations and solve them
(using F4 CAL LINEQS), correct the current values of the para-
meters Bj, check to see whether the new values of the Bj result
in a smaller sum of squares, and, if so, iterate the procedure,
in the event that the new values of the parameters fail to give
a smaller sum of squares, the parameters are reset to values
which give the minimum thus-far-discovered sum of squares, are
corrected by scaled-down amounts of the original corrections, and
the sum of squares is again tested.
These iterations proceed until one of four possible terminating
conditions occurs. GAUSS will then return control to the calling
program with the flag. CNTROL(8), set to indicate the terminating
condition:
CNTROL(8) = 0
CNTROL(8) = 1
193
Condi tion
If convergence (a change of less than 10"5)
of the values of the parameters B. occurs.
J
If the Bj iterations proceed until the scaled
down corrections to the parameters become
negligible without producing a new minimum in
the sum of squares.
-------
CNTROL(8) =2 If subroutine GAUSS finds that the normal
equations cannot be solved. The offending
matrices are recorded.
CNTROL(8) =3 If the number of iterations exceeds LIMIT
or CNTROL(4).
VI. Subroutine FINALE
FINALE is invoked by the call
CALL FINALE(CNTROL,B,Z,MDIM,NDIM)
where all arguments are the same as those used for the GAUSS call.
The subroutine, when supplied with the indicated arguments, gives a
tabulation of results. FINALE first gives a tabulation of the values
of the dependent variable (i.e., Zj or Y OBSERVED as it is referred
to, in the output) for the totality of data points. Corresponding
to each of these values is the corresponding value of the fitting
function, evaluated for the supplied B and Z arrays, the difference
between these two values, and the percent deviation of the observed
value from the calculated value of the dependent variable.
Then there follows a listing of the average and maximum deviation.
It may be noted that, following the values for the maximum devia-
tion and the maximum percent deviation, there are two integer entries.
These are the indices of the data points which gave rise to these
maximum deviations. Often such points represent errors in the data
and it may be worthwhile to repeat the analysis with the offending
points corrected or deleted.
VII. Function DERIV
This subprogram, FUNCTION DERIV(K,N,B,Z,BDIM), supplies the partial
derivative of YCOMP with respect to the variable B(K), evaluated at
the n-th data point. The version supplied with the program computes
the value required by a symmetric differencing scheme. However, the
user may always compile his own version to supply the exact partial
derivatives. In small test problems, the final values obtained for
the parameters are accurate to at least six significant figures; so
the general purpose version of DERIV will be adequate for most
problems.
VIII. Other Subroutines
A. Subroutine INDATA, invoked by the call
CALL INDATA(CNTROL, B,Z,MDIM,NDIM,BDIM)
194
-------
reads in the data and prints titles.
B. Subroutine BTEST, invoked by the call
CALL BTEST (CNTROL,B.BDIM)
Tests the amount of change in the parameters B.
C. Subroutine DMPMAT, invoked by the call
CALL DMPMAT(CNTROL,BDIM,A,B(1,3))
prints intermediate iterations according to the user's speci-
fications in CNTROL(5).
Subroutines Needed:
User's own YCOMP
TIMING
The sample problem used 3.226.seconds of central processor time and
8.712 seconds of peripheral processor time. See the writeup for G2 BC
NLIN for comparisons of the time and accuracy yielded in using 6 B
parameters and in using 4 B parameters (a simplified equation).
SIZE
DMPMAT
GAUSS
LINEQF
Embedding routine
INDATA
SETUP
FINALE
DERIV
33168locations
6018locations
372elocations
4648locations
4548locations
141glocations
1208locations
3068locations
195
-------
APPENDIX C
BATCH CULTURE ASSAY RESULTS
TABLE C-I. FIRST INTERLABORATORY PRECISION TEST
100% PAAP
Flask
A1
Flask A2
Flask A3
Flask
A4
Flask
A5
Day
oH
CC
SS
CV
pH
CC
SS
CV
dH
CC
SS
CV
pH
CC
SS
CV
pH
CC
SS
CV
0
7.0
0.001
-
7.0
0.001
.
7.0
0.001
_
7.0
0.001
_
7.0
0.001
3
7.4
0.007
.43
/.b
0.006
.29
7. 7
0.007
. 37
7.8
0.004
.22
7.7
0.011
.59
5
7.6
O.OOA
.48
7.9
0.009
.49
7.9
0.014
.80
7.9
0.005
.28
7.9
0.072
3.68
7
7.6
0.035
1.98
/. /
0.031
1.81
7.8
0.108
5.49
7.7
0.008
.42
8.9
0.413
29.5
9
8.0
0. 312
16.54
/. 9
0.069
3.26
9.2
0.550
37.5
8.7
0.012
.81
10.7
1 .511
124. 7
1 3
10.4
1.985
121,1
8,8
0,334
17.5
10.4
2.672
126.9
8.6
0.023
1.27
10.2
1.656
132.5
l 7
9.7
3. 708
F
8.8
1.698
F
9.8
4.209
F
8.9
0.017
F
9.9
3.826
F
21
y.s
5.057
2bb
F
9.6
3.435
175
F
9.2
7.098
300
F
8.5
0.043
8.6
F
9.2
5.856
268.5
F
30% PAAP
Mask
R1
Flask B2
Flask
B3
Flask
B4
Flask
B5
Day
pH
CC
SS
CV
PH
CC SS
CV
pH
CC
SS
CV
pH
CC
SS
CV
pH
CC
SS
CV
0
7.0
0.001
.
7.0
0.001
_
7.0
0.001
_
7.0
0.001
_
7.0
0.001
3
/. 4
0.003
.11
/. b
0.004
.18
7.7
0.004
.u
7.8
0.001
0.05
7.7
0.001
0.05
5
7.9
0.002
.10
/. 9
0.001
.04
7.8
0.004
.22
7.8
0.001
0.05
7.9
0.001
0.05
7
y.j
0.005
.27
/. 4
0.005
.29
7.5
0.002
.08
7.6
0.002
0.07
7.5
0.001
0.07
9
7.9
0.004
.26
/. 9
0.002
.19
7.7
0.004
.23
7.8
0.001
0.07
7.9
0.001
0.02
1 3
8.2
0.022
2.05
8.5
0.003
.22
8. 3
0.006
. 39
8.0
0.003
0.23
8.0
A
A
17
8.4
0.007
F
8.2
0.008
F
8.1
0.010
F
8.0
0.002
F
7.8
0.001
F
21
8.3
0.011
B
F
8.1
0.249 13.2
F
7.9
0.058
2.0
F
7. 7
0.015
B
F
7.4
0.002
B
F
10% PAAP
Flask
CI
Flask
C2
Flask C3
Flask C4
Flask
C5
Day
pH
CC
SS
CV
pH
CC
SS
CV
PH
CC SS
CV
pH
CC SS
a
pH
CC
SS
CV
0
7.0
0.001
7.0
.
7.0
0.001
_
7.0
0.001
.
7.0
0.001
_
3
7.4
0.0007
0.03
7.7
0.04
7.4
0.0004
0.01
7.4
0.0004
0.02
7.4
0.0009
0.04
5
7.6
0.0007
0.03
7.2
0.01
7.3
0.0005
0.05
7.3
0.000 3
0.02
7.2
0.0002
0.01
7
7. 7
0.0002
0.00
7.6
0.02
7.5
0.0005
0.05
7.4
0.0004
0.01
7.3
0.0002
O.OC
9
7.8
0.0003
0.02
7.8
0.03
7.6
0.0002
0.01
7.3
0.0002
0.01
7.6
0.0002
0.01
13
8.1
0.0003
0.02
7.9
0.001
0.02
7.9
0.0003
0.01
7.8
0.0002
o.oc
7. 7
0.0003
0.03
1 7
7.8
0.0002
F
7.7
0.002
F
7.6
0.0002
F
7.6
0.0003
F
7.5
0.0002
F
21
7.9
0.0004
B
F
7.7
0.002
B
F
7.8
0.0002 B
F
7.2
0.0002 B
F
7.8
0.0002
B
F
1% PAAP
Flask
D1
Flask D2
Flask 03
Flask D4
Flask
D5
Oay
pH
CC
SS
CV
pH
CC SS
CV
PH
CC SS
CV
pH
CC SS
CV
pH
CC
SS
CV
0
7.0
0.001
_
7.0
0.001
.
7.0
0.001
7.0
0.001
.
7.0
0.001
_
3
7.6
0.008
0.03
7.6
0.0004
0.01
7.5
0.0003
0.01
7.2
0.0007
0.03
7.0
0.0004
C.01
5
7.4
0.0004
0.01
7.6
0.0004
0.02
7.5
0.0001
0.00
7.2
0.0003
0.02
7.2
0.0002
0.00
7
7.6
0.0002
O.Ol
7.6
0.0002
0.01
7.3
0.0001
0.00
7.2
0.0002
0.00
7.2
0.0001
0.00
9
7. 5
0.0002
0.01
7.4
0.0003
0.02
7.3
0.0001
0.00
7.2
0.0002
0.01
7.2
0.0001
0.01
13
7.5
0.0002
0.02
7.6
0.0002
0.01
7.5
0.0001
0.01
7.4
0.0003
0.04
7.3
0.0001
o.oc
1 7
7.5
0.0003
F
7.5
0.0002
F
7.4
0.0001
F
7.2
0.0001
F
7.1
0.0002
F
21
7.5
0.0003
B
F
7.5
0.0002 B
F
7.0
0.0002 B
F
7.0
0.0002 B
F
6.9
0.0001
B
F
A - sooiled
B - too low to measure
CC - cell count x 10f>/mt
SS - mq suspended solids/?.
CV —cell volume concentration, 106u3/im
F - instrument in repair
196
-------
APPENDIX C
TABLE C—II. SECOND INTERLABORATORY PRECISION TEST
100% PAAP
Flask A1
Flask A2
Flask A3
Flask A4
Flask A5
o
><
dH
CC
SS
TCV
pH
CC
SS
TCV
PH
CC
SS
TCV
PH
CC
SS
TCV
pH
CC SS
TCV
0
7,0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
_
3
7.0
0.021
F
6.9
0.007
F
6.9
0.006
F
7.0
0.003
F
7.0
0.005
F
5
7.8
0.459
40.88
8.0
0.547
52.8
7.9
0.398
35.66
8.0
0.045
4.28
8.0
0.267
22.93
7
8.2
3.272
242.11
8.0
2.124
159.3
8.2
2.82
229.16
8.0
0.071
6.53
7.9
1.124
2.29
9
8,6
9,909
678,8
8.7
12.627
1111.17
9.0
9.64
72.96
8.0
0.124
11.33
8.7
7.761
539.42
13
8.1
48. 700
3068.08
8.2
19.353
1219.26
8.1
20.25
1306.0
8.1
0.321
26.15
8.3
22.395
1567.64
17
8.3
27.773
1642.7
8.5
31.227
1904.8
8.5
35.64
2174.1
8.4
0.193
17.05
8.6
27.324
1352.6
21
32.508
381.3
2048.0
29.058
377.5
1874.0
27. 37
400.0
1582.0
0.188
6.8
18.986
35.226 350.0
2131.2
30% PAAP
Flask B1
Flask B2
Flask B3
Flask B4
Flask B5
Day
pH
CC SS
TCV
PH
CC
SS
TCV
PH
CC
SS
TCV
pH
CC
SS
TCV
pH
CC
SS
TCV
0
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
3
7.0
0.006
F
6.9
0.007
F
6.9
0.010
F
6.6
0.015
F
6. 7
0.008
F
5
8.1
0.176
18.69
8.0
0.679
62.46
7.8
1.22
129,94
7,3
0,572
54.96
7.3
0.544
53.07
7
7.7
0.666
59.9
7. 7
4.938
335.8
7.9
4.99
334.60
7.7
4.132
276.87
7.7
3.555
261.27
9
8.5
3.357
288.68
8.5
5.524
383.9
8.2
6.87
470.96
8.3
5.574
557.43
7.7
7.753
542.71
13
8.0
8.463
842.05
8.0
7.195
496.48
7.7
6.67
437.37
7.7
6.952
472.7
7.6
7.029
492.5
17
8.4
6.370
423.62
8.2
6.606
495.9
8.2
8.52
549.44
8.1
9.535
524.55
8.1
8.93
669.24
21
7,485 158.0
523,9
7.137
178.0
478.2
8.96
153.5
586.8
9.859
162.5
501.39
8.58
162.0
609.24
10%
PAAP
Flask CI
Flask C2
Flask C3
Flask C4
Flask C5
Day
pH
CC SS
TCV
pH
CC
SS
TCV
pH
CC
SS
TCV
pH
CC
SS
TCV
pH
CC
SS
TCV
0
7.0
0.001
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
3
6.7
0.002
F
6.7
0.004
F
6.8
0.004
F
6.1
0.002
F
6.0
0.005
F
5
7.2
0.018
2.27
7.8
0.016
2.52
7.7
0.004
0.398
7.7
0.007
0.B58
7.6
0.202
2.02
7
7.7
0.022
1.98
7.5
0.057
5.66
7.3
0.007
0.745
7.2
O.O&fi
0.888
7.2
0.811
66.49
9
7.6
0.026
2.71
7.7
0.233
24.17
7.3
o.on
0.99
7.4
0.016
1.88
7.0
2.091
150.58
13
8.0
0.036
3.136
8.0
2.69
211.18
7.3
0.021
1.585
7.1
0.019
2.06
7.3
3.546
239 . 34
17
7.6
0.039
2.953
7.5
1.783
141.7
7.3
0.016
1.151
7.2
0.021
1.121
7.3
2.534
200.59
21
0.012 1.83
1.209
1.729
47.7
159.1
0.008
1.55
0.764
0.021
15.8
2.345
2.524
62.9
163.4
H PAAP
Flask D1
Flask 02
Flask D3
Flask D4
Flask 05
Day
pH
CC SS
TCV
PH
CC
SS
TCV
pH
CC
SS
TCV
PH
CC
SS
TCV
pH
CC
SS
TCV
0
7.0
0.001
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
7.0
0.001
-
3
5.8
0.0002
5.6
0.0005
F
5.6
0.0001
F
5.6
0.0003
F
5.4
0.0002
F
5
7.1
0.0024
0.304
7.5
0.0019
0.268
7.2
0.0007
0.086
7.0
0.0016
0.137
6.9
0.001
0.125
7
7.2
0.0124
1.377
7.0
0.0034
0.395
7.1
A
F
6.9
A
F
6.8
A
F
9
7.1
A
1.275
7.1
0.0016
0.198
7.1
0.0022
0.286
6.9
0.0024
0.269
6.9
0.0024
0.299
13
7.2
0.0042
0.363
7.1
0.0103
1.073
7.0
0.0033
0.197
6.9
0.0045
0.286
6.8
0.0048
0.261
17
7.4
0.0046
1.363
7.3
0.0018
0.379
7.2
0.0064
0.334
7.0
0.0049
0.395
6.9
0.0029
0.132
21
0.0040 B
0.633
0.0018
B
0.216
0.0064
B
0.838
0.0049
B
0.578
0.0029
B
0.316
A - spoiled
B — too low to measure
CC — cell count x 106/mt
TCV— cell volume concentration x 10fiyVm*
F - Instrument In repair
197
-------
APPENDIX C
TABLE C-III. CLEAR LAKE STUDY
100% Clear Lake 100% Clear Lake 100% Clear Lake
Day
pH
ccb
TCVc
ssd
pH
cc
TCV
SS
pH
CC
TCV
SS
0
0.052
4.6
0.057
5.5
0.093
10.3
2
8.0
0.114
14.3
7.9
0.102
12.5
8.0
0.107
13.7
3
8.4
0.118
15.2
8.4
0.133
16.8
8.4
0.113
15.9
4
8.1
0.394
38.0
8.1
0.407
38.7
8.1
0.429
41.0
5
7.1
0.479
42.7
7.9
0.490
44.1
7.8
0.506
45.5
6
7.9
0.467
40.4
8.1
0.497
42.5
8.3
0.512
44.3
7
8.1
0.704
72.5
8.2
0.506
46.5
8.4
0.565
53.1
10
6.9
0.541
47.6
41.3
7.9
0.562
48.3
36.4
8.2
0.616
55.7
39.6
66.It Clear Lake3
66.7% Clear Lake
66.7% Clear Lake
Day
pH
cc
TCV
ss
pH
cc
TCV
SS
pH
CC
TCV
SS
0
0.078
8.5
0.073
7.9
0.094
10.1
2
7.8
0.054
6.9
7.8
0.078
10.0
7.8
0.082
10.9
3
8.3
0.088
11.3
8.2
0.125
15.6
8.2
0.093
12.1
4
8.1
0.295
26.1
8.1
0.311
26.6
8.1
0. 303
27.2
5
7.9
0.348
29.1
8.0
0.416
38.4
8.0
0.406
36.3
6
8.3
0.381
30. 7
8.2
0.375
29.4
8.3
0.393
38.0
7
8.2
0.472
40.6
8.2
0.426
35.2
8.1
0.494
50.6
10
8.1
0.443
36.8
30.0
7.9
0.477
39.3
28.0
7.9
0.436
35.6
24.1
33.3% Clear Lake3
33. 3% Clear Lake
33. 3% Clear Lake
Day
pH
cc
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.022
2.2
0.019
1.7
0.020
1.8
2
7.9
0.033
4.8
7.9
0.027
4.1
8.0
0.027
4.0
3
8.2
0.080
8.6
8.1
0.070
7.9
8.0
0.066
7.4
4
8.1
0.131
13.5
8.1
0.123
12.0
8.1
0.122
12.6
5
7.8
0.189
17.4
7.6
0.168
14.8
7.5
0.190
18.3
6
8.2
0.335
20.0
8.3
0.232
20.3
8.3
0.216
17.8
7
7.9
0.285
28.9
7.8
0.232
19.1
7.7
0.252
20.9
10
6.5
0.356
28.4
28.2
6.9
0.306
24.7
25.0
7.6
0.310
25.0
30% PAAP
Control
30% PAAP
Control
30% PAAP
Control
Day
PH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.010
0.6
0.009
0.6
0.010
0.7
2
7.5
0.018
2.6
7.5
0.019
2.5
7.6
0.015
1.6
3
8.2
0.053
5.1
8.1
0.054
5.0
8.0
0.025
2.9
4
8.6
1.028
76.6
7.5
1.121
80.7
7.6
0.822
56.3
5
7.8
2.328
178.1
7.5
2.749
233.6
7.9
1.566
143.3
6
8.2
4.318
405.9
8.3
4.567
408.8
7.6
3.214
287.7
7
8.1
6.155
538.6
7.9
6.192
504.6
7.9
6.076
607.6
10
7.5
7.216
570.1
152.6
7.4
7.006
514.9
151.8
7.7
7.081
501.0
140.5
Clear Lake + 9% PAAP
Clear Lake
+ 9% PAAP
Clear Lake + 9% PAAP
Day
PH
CC
TCV
SS
PH
CC
TCV
SS
pH
CC
TCV
SS
0
0.123
12.8
0.110
12.3
2
7.8
0.074
9.5
7.8
0.085
10.8
7.9
0.100
12.3
3
8.1
0.136
18.2
8.3
0.087
12.8
8.4
0.121
15.1
4
8.1
1.149
116.6
8.2
0.934
94.8
8.2
1.117
98.8
5
7.9
3.153
268.0
7.7
2.968
265.5
7.8
2.580
212.9
6
1.7
3.264
290.5
7.9
3.270
264.9
8.1
3.574
311.0
7
8.1
3.519
299.1
8.1
3.434
291.9
8.2
3.754
309.7
10
8.0
3.884
306.8
111.8
7.7
3.826
298.5
94.4
7.6
3.887
308.1
101.3
aDi1uted with demineralized distilled water. cTotal cell volume (p3 x 106/mi)
bCell counts with 106/m«. Suspended solids (mg/O
198
-------
APPENDIX C
TABLE C—IV. SECONDARY EFFLUENT ENRICHMENT OF CLEAR LAKE WATER
10% Secondary Effluent3 10% Secondary Effluent 10% Secondary Effluent
Day
PH
CCb
TCVc
SSd
pH
CC
TCV
SS
PH
CC
TCV
SS
0
0.043
5.0
0.047
5.2
0.044
4.3
2
8.0
0.040
6.1
8.0
0.032
4.8
7.9
0.045
6.9
3
8.3
0.054
8.4
8.3
0.056
8.5
8.4
0.054
8.2
4
8.1
1.177
120.6
8.1
1.330
139.0
8.1
1.100
103.3
5
8.1
2.653
229.5
8.3
2.991
270.7
8.4
2.799
254.7
6
8.4
3.888
316.9
8.2
3.767
325.8
8.2
3.821
326.7
7
8.3
4.589
449.7
8.3
4.401
404.8
8.3
4.148
392.1
10
7.9
4.528
352.1
129.1
7.5
4.517
356.8
146.1
7.5
4.382
346.0
116.1
20% Secondary Effluent
20% Secondary Effluent
20% Secondary Effluent
Day
dH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.046
5.2
0.039
4.6
0.045
4.9
2
7.9
0.054
7.9
8.0
0.052
8.3
8.0
0.041
6.1
3
8.3
0.064
10.4
8.4
0.059
10.1
8.5
0.047
7.4
4
8.1
1.660
188.4
8.1
1.240
117.1
8.2
0.836
62.7
5
8.3
2.490
252.2
8.4
3.049
309.5
8.2
0.824
82.4
6
8.1
4.651
434.8
8.5
5.900
548.7
8.2
1.712
126.7
7
8.4
6.216
618.5
8.4
7.213
692.5
8.4
1.364
137.1
10
8.0
8.148
900.7
205.7
7.9
7.995
627.5
198.3
8.2
1.472
175.9
141.3
30% Secondary Effluent
30% Secondary Effluent
30% Secondary Effluent
Day
pH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.043
4.6
0.048
5.3
0.044
4.7
2
7.9
0.063
9.4
7.8
0.065
9.6
7.8
0.059
8.8
3
8.4
0.082
13.3
8.4
0.080
12.8
8.4
0.093
14.6
4
8.2
1.466
157.6
8.1
1.475
150.4
8.1
1.326
130.6
5
8.6
3.203
326.7
8.4
2.548
286.6
8.5
2.449
262.1
6
8.3
6.096
542.5
8.5
5.274
527.4
8.5
6.122
547.8
7
8.5
12.562
1249.9
8.6
10.853
1079.0
8.6
11.107
1089.6
10
8.1
9.489
865.9
266.3
8.0
9.867
821.4
242.1
7.7
9.208
752.8
309.0
aDi 1 uted with Clear Lake water. cTotal cell volume (106p3/mJt)
bCel 1 counts x 106M Suspended solids (mg/i)
199
-------
APPENDIX C
TABLE C-V. CLEAR LAKE ENRICHMENT (SPIKING) TESTS3
CI + P CI + Fe CI + Fe
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.366 39.6
8.3 0.578 61.9
0.414 38.1
0.792 77.2
1.551 161.3
1.459 135.7
1.557 127.1
2.811 213.6 87.6
0.345 36.2
8.3 0.669 70.9
0.429 38.2
0.570 54.4
1.231 124.3
1.145 97.3
1.294 99.6
2.189 166.4 97.2
0.199 20.3
8.3 0.937 83.4
0.395
0.744 68.5
0.888 86.9
0.720 65.5
2.925 228.1 100.0
CI + N, P CI + N, P CI + N, Fe
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.340 35.7
8.4 0.383 36.5
0.385 36.9
0.599 57.2
0.818 86.7
0.780 78.0
1.879 176.6
5.689 449.4 100.0
0.273 26.5
8.3 0.564 53.0
0.358 36.9
0.580 62.6
0.705 71.9
0.718 73.9
1.386 115.0
4.867 384.5 122.0
0.238 23.1
8.3 0.662 68.9
0.703 70.3
0.663 76.9
0.948 94.8
0.804 74.8
2.701 210.7
4.190 360.3 191.5
CI + N, Fe CI + P, Fe CI + P, Fe
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.308 35.1
8.3 0.496 52.1
0.468 49.1
0.476 46.2
0.840 86.6
0.819 82.8
1.326 104.7
3.224 280.5 122.4
0.320 33.9
8.3 0.563 52.9
0.475 42.8
0.464 41.8
0.847 80.4
1.198 115.1
1.218 90.1
3.005 234.4 100.0
0.284 29.5
8.3 0.493 50.3
0.424 38.6
0.812 80.8
1.889 213.5
1.019 95.7
1.140 78.7
2.594 199.7 104.0
CI + N, P. Fe CI + N, P, Fe CI + N, P. Fe, Micron.
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.315 34.3
8.3 0.625 61.9
0.570 53.1
0.762 78.2
1.149 125.3
0.924 81.3
2.501 200.0
5.033 427.8 144.2
0.285 30.2
8.3 0.422 46.0
0.416 43.5
0.570 62.8
0.804 80.4
1.212 120.0
1.726 157.0
4.389 373.1 126.5
0.301 31.9
8.4 0.687 63.2
0.998 83.1
1.527 132.9
2.329 193.3
2.534 205.2
4 . 319 306 .6
9.056 724.5 202.5
200
-------
TABLE C-V (Continued)
CI + N, P, Fe, Micron. CI + 30% PAAP CI + 30% PAAP
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.351 42.1
0.328 38.0
0.319 36.7
8.4 0.815 76.6
8.5 1.034 109.6
8.5 0.865 88.3
1.237 116.5
0.906 83.4
1.249 138.6
2.061 206.1
2.128 195.8
1.717 195.4
2.678 227.6
2.150 195.6
2.295 257.1
3.099 266.5
3.310 284.7
2.089 206.0
5.067 476.3
6.705 509.5
6.248 556.0
13.100 1558.0 212.5
12.604 1499.8 300.0
15.759 2096.0 310.4
100% CI 100% CI 100% CI
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.301 32.5
8.3 0.358 36.2
0.506 53.4
0.118 12.1
0.826 83.4
1.174 108.0
2.000 192.0
2.870 236.1 160.0
0.320 34.9
8.3 0.381 41.2
0.402 43.2
0.466 45.5
0.819 81.1
1.109 108.7
1.211 104.1
2.732 225.4 108.0
0.283 30.3
8.3 0.421 42.5
0.567 59.5
0.485 49.2
0.936 95.4
0.964 98.3
3.104 310.4
2.466 204.7 94.8
CI + 10% SE CI + 10% SE CI + 10% SE
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.267 29.9
8.3 0.419 45.7
0.394 42.3
0.602 63.5
0.737 78.1
0.936 111.4
2.552 244.9
5.220 433.3 127.4
0.290 32.2
8.3 0.464 53.1
0.411 43.1
0.464 51.5
0.547 62.3
1.043 119.9
1.646 182.7
4.572 420.6 124.5
0.297 34.2
8.3 0.422 44.7
0.509 50.4
0.632 67.4
1.155 135.1
1.285 142.7
2.205 194.1
5.784 497.4 164.7
CI + N CI + N CI + P
pH CC TCV SS
pH CC TCV SS
pH CC TCV SS
0.343 36.9
8.3 0.400 42.4
0.415 41.9
0.706 72.4
0.832 91.6
1.204 119.2
2.410 226.5
3.371 330.4 140.0
0.314 32.1
8.3 0.426 42.6
0.297 31.7
0.480 50.2
1.123 131.4
1.022 122.6
1.044 93.9
2.754 250.6 92.3
0.334 35.4
8.3 0.380 38.6
0.490 45.8
0.893 99.6
1.654 168.7
0.890 83.7
2.268 199.6
2.472 187.9 98.7
Measurements made on days 2, 3, 4, 5, 6, 7, 10, 16, respectively.
-------
APPENDIX C
TABLE C-VI. SACRAMENTO RIVER STUDY
100% SRa
100% SR
100% SR
Day
PH
CCb
TCVC
SSd
PH
CC
TCV
SS
PH
CC
TCV
SS
0
0.030
3.2
0.034
3.6
0.034
3.3
3
8.0
0.065
8.4
8.9
0.074
9.1
7.9
0.074
8.4
4
0.212
16.8
0.213
20.9
0.298
25.4
6
0.238
19.9
0.247
21.5
0.260
18.5
7
0.276
20.7
0.274
23.8
0.301
23.5
10
0.363
26.5
0.284
23.6
0.283
22.3
14
0.389
32.3
15.6
0.311
25.7
17.1
0.301
21.7
13.0
66
7% SR
66
7i SR
66
7% SR
Day
PH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.042
2.8
0.025
2.1
0.026
2.3
3
7.9
0.064
6.3
7.9
0.102
8.6
8.3
0.115
9.7
4
0.192
18.4
0.222
19.6
0.179
12.7
6
0.193
13.6
0.195
12.9
0.208
14.5
7
0.214
15.1
0.216
15.3
0.215
15.0
10
0.258
18.6
0.256
18.8
0.262
18.4
14
0.307
24.6
10.0
0.229
19.8
10.3
0.237
15.6
10.8
33.3% SR
33.3% SR
33.
3% SR
Day
PH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
0.019
1.2
0.018
1.1
0.019
1.7
3
7.5
0.067
4.7
7.6
0.065
4.7
7.5
0.069
5.2
4
0.087
6.6
0.087
7.1
0.085
6.1
6
0.092
6.0
0.094
6.4
0.095
6.4
7
0.103
6.8
0.105
7.1
0.107
7.5
10
0.113
7.9
0.111
7.4
0.123
8.6
14
0.113
7.8
6.0
0.130
9.0
5.8
0.159
11.3
5.7
30% PAAP Control
30% PAAP Control
30% PAAP Control
Day
PH
CC
TCV
SS
PH
CC
TCV
SS
pH
CC
TCV
SS
0
0.022
1.6
0.020
1.3
0.022
1.4
3
8.4
0.075
8.4
7.1
0.081
8.3
6.9
0.061
6.2
4
0.378
39.3
0.301
33.5
0.238
22.2
6
3.882
287.2
2.031
204.0
2.895
266.4
7
8.6
4.848
360.6
8.2
3.232
302.2
8.1
5.532
409.4
10
5.936
463.0
3.945
374.8
7.447
536.2
14
6.649
558.5
157.6
3.810
382.9
139.9
5.296
389.2
159.1
SR +
30% PAAP
SR +
30% PAAP
SR +
30% PAAP
Day
pH
CC
TCV
SS
PH
CC
TCV
SS
PH
CC
TCV
SS
0
0.026
1.9
0.060
6.1
0.030
3.1
3
7.0
0.040
5.6
7.4
0.054
8.4
7.5
0.071
11.6
4
1.959
25.3
0.268
32.4
0.197
24.2
fi
2.330
188.8
5.197
462.5
2.797
243.4
7
8.1
2.773
196.8
8.0
2.927
209.3
8.1
4.098
293.0
10
4.214
598.4
3.703
277.7
3.654
358.0
14
4.942
375.6
70.7
3.933
298.9
64.0
4.408
335.0
70.0
aSacramento River water. cTotal cell volume (10VM)
bCel 1 counts x 106/m«. dSuspended Solids (mg/O
202
-------
TABLE C—VII.
APPENDIX C
SECONDARY EFFLUENT ENRICHMENT OF SACRAMENTO RIVER WATER
+10% SEa +10% SE +10% SE
Day
dH
CCb
TCVC
SSd
PH
CC
TCV
SS
PH
CC
TCV
SS
0
0.025
2.1
0.027
2.2
0.029
2.7
3
8.1
0.073
11.4
7.8
0.059
9.0
8.0
0.051
7.3
4
0.422
50.6
0.244
31.3
0.229
35.5
6
2.804
243.3
2.700
199.8
3.302
174.0
7
8.5
3.773
317.0
8.5
3.878
312.2
8.5
4.382
339.6
10
3.648
277.2
3.562
277.8
5.621
455.3
14
5.285
422.8
73.1
8.901
716.5
91.5
7.198
529.0
115.8
+20% SE
+20% SE
+20% SE
Day
PH
CC
TCV
SS
pH
CC
TCV
SS
PH
CC
TCV
SS
0
0.026
2.6
0.035
3.7
0.030
2.8
3
7.8
0.088
14.8
8.5
0.097
14.9
7.3
0.080
9.9
4
0.383
59.8
0.332
49.4
0.297
38.3
6
5.140
439.5
3.717
306.7
3.499
265.9
7
8.5
6.218
500.6
8.5
5.050
414.1
8.5
4.942
390.1
10
8.264
673.5
5.444
454.6
4.947
405.7
14
7.477
568.2
165.2
8.087
622.7
131.9
5.518
444.2
134.5
+30% SE
+30% SE
+30% SE
Day
pH
CC
TCV
SS
pH
CC
TCV
SS
PH
CC
TCV
SS
0
0.042
4.4
0.035
3.4
0.038
4.3
3
7.6
0.060
9.7
7.8
0.063
6.0
8.2
0.061
9.8
4
0.288
48.1
0.209
34.1
0.195
30.8
6
5.204
463.4
2.853
286.7
3.452
329.6
7
8.6
8.067
705.8
8.6
6.282
559.1
8.2
6.463
568.8
10
10.146
842.1
9.766
942.4
8.814
696.3
14
14.428
1154.2
247.0
13.675
1053.0
221.5
13.823
1089.9
210.4
Secondary Effluent. °Total cell volume (106u3/mi.)
bCel1 counts x 106/nw Suspended solids (mgA)
203
-------
TABLE C—VIII.
APPENDIX C
SACRAMENTO RIVER ENRICHMENT (SPIKING) TESTS
100% SR 100% SR 100% SR
Day
pH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
8.45
0.099
12.3
8.45
0.099
12.4
8.45
0.103
12.7
2
0.128
14.0
0.065
7.9
0.097
10.9
3
8.45
0.116
11.9
8.35
0.139
14.6
8.45
0.147
16.9
4
8.2
0.266
27.4
8.3
0.243
23.6
8.2
0.230
23.5
5
0.171
16.8
0.193
18.5
0.204
20.8
6
0.215
20.5
0.238
22.9
0.272
26.1
7
0.255
22.8
0.250
21.7
0.259
22.6
10
0.942
65.2
32.8
0.860
60.2
30.4
0.955
71.7
31.7
SR +
10% SE
SR +
10% SE
SR +
10% SE
Day
OH
CC
TCV
SS
PH
CC
TCV
SS
pH
CC
TCV
SS
0
8.4
0.108
13.3
8.4
0.129
15.5
8.4
0.141
17.0
2
0.068
9.3
0.092
12.1
0.072
9.9
3
8.35
0.146
20.0
8.55
0.142
19.3
8.55
0.139
20.4
4
8.5
0.261
31.4
8.3
0.188
25.7
8.4
0.209
27.6
5
0.229
27.5
0.226
29.2
0.254
24.6
6
0.262
30.2
0.322
39.6
0.286
35.5
7
0.962
75.5
1.217
124.1
1.465
161.9
10
4.189
332.0
73.6
4.131
313.9
72.2
1.962
151.1
80.0
SR + N
SR
+ N
SR
+ P
Day
pH
CC
TCV
SS
pH
CC
TCV
SS
pH
CC
TCV
SS
0
8.45
0.090
n.2
8.45
0.083
11.1
8.45
0.084
11.1
2
0.116
13.3
0.111
12.8
0.107
12.4
3
8.4
0.148
15.9
8.6
0.156
16.8
8.6
0.168
17.5
4
8.4
0.163
16.6
8.4
0.151
15.3
8.4
0.183
18.5
5
0.189
18.3
0.155
13.9
0.209
20.2
6
0.209
19.9
0.231
21.2
0.249
23.0
7
0.242
21.0
0.247
21.7
0.304
28.5
10
1.391
114.8
43.1
4.450
318.2
45.1
1.564
122.4
58.7
SR
+ P
SR
+ Fe
SR
+ Fe
Day
PH
CC
TCV
SS
PH
CC
TCV
SS
pH
CC
TCV
SS
0
8.45
0.106
11.8
8.45
0.080
9.4
8.45
0.154
14.3
2
0.110
12.4
0.138
15.3
0.146
15.3
3
8.5
0.160
15.4
8.55
0.149
14.6
8.35
0.174
17.9
4
8.4
0.152
14.3
8.4
0.162
15.0
8.4
0.152
14.3
5
0.175
16.9
0.939
77.4
0.305
107.7
6
0.283
28.6
1.149
92.5
0.447
38.9
7
0.271
23.6
0.830
63.5
0.963
73.2
10
0.984
75.5
37.8
0.997
74.8
34.1
1.063
77.6
32.3
SR +
N, P
SR +
N, P
SR +
N, Fe
Day
dh
CC
TCV
SS
pH
CC
TCV
SS
PH
CC
TCV
SS
0
8.45
0.078
9.1
8.45
0.130
13.3
8.45
0.093
11.3
2
0.105
12.2
0.126
12.6
0.133
14.7
3
8.5
0.220
20.8
8.45
0.262
24.7
8.45
0.202
20.4
4
8.5
0.229
20.4
8.4
0.265
22.8
8.4
0.289
28.3
5
1.083
91.5
1.302
113.9
1.341
104.6
6
1.382
106.4
2.247
174.1
1.636
111.2
7
1.960
149.0
3.285
236.5
4.889
364.0
10
7.602
447.0
111.6
7.861
495.2
110.5
2.200
184.8
68.6
20b
-------
lay
0
2
3
4
5
6
7
0
iayj"
0
2
3
4
5
6
7
0_
lay
0
2
3
4
5
6
7
0
lay
0
2
3
4
5
6
7
0
TABLE C—VI11 (Continued)
SR + N, Fe
SR + P, Fe
SR + P, Fe
pH
CC
TCV
ss
pH
CC
TCV
SS
pH
CC
TCV
SS
8.45
8.45
8.7
0.092
0.112
0.218
0.259
1.155
1.576
1.374
1.674
11.3
12.6
21.9
25.4
97.6
110.3
93.4
146.1
8.45
8.5
8.3
54.3
0.100
0.099
0.193
0.274
0.850
1.231
1.387
1.044
11.9
11.2
17.9
25.5
67.5
82.5
101.2
77.2
8.45
8.45
8.5
35.6
0.066
0.097
0.196
0.253
2.755
1.414
2.408
1.012
7.6
10.4
18.8
21.7
181.8
131.6
169.8
79.7
33.3
SR + N, P, Fe
SR + N, P, Fe
SR + N, P, Fe, Micron.
PH
CC
TCV
SS
PH
CC
TCV
SS
PH
CC
TCV
SS
8.45
8.4
8.4
0.088
0.113
0.196
0.237
0.983
1.594
1.844
7.671
11.0
12.4
18.8
22.3
87.0
121.9
131.8
487.1
8.0
8.5
8.4
119.5
0.076
0.124
0.222
0.279
1.061
1.732
2.634
8.102
9.4
13.5
21.8
26.2
91.8
129.9
183,1
514.5 110.6
8.45
8.6
8.7
0.093
0.160
0.496
4.961
6.516
6.886
6.262
8.476
11.2
18.1
47.9
352.2
436.6
454.5
388.3
529.7
126.2
SR + N, P, Fe, Micron.
SR + 30% PAAP
SR + 30% PAAP
nH
CC
TCV
SS
PH
CC
TCV
SS
PH
CC
TCV
SS
8.45
8.25
8.8
0.115
0.142
0.441
2.528
5.330
7.694
8.249
9.036
11.9
15.8
45.5
203.5
378.4
480.8
519.7
560.2 150.0
8.0
8.35
8.7
0.060
0.084
0.229
2.430
4.822
7.234
8.223
10.084
4.1
8.6
21.6
240.5
361.6
491.8
542.7
675.6
8.1
8.35
8.7
182.6
0.087
0.144
0.357
1.913
2.856
6.087
7.463
10.240
8.4
15.8
36.2
156.9
217.0
404.8
548.5
701.4
182.6
30% PAAP
PH
CC
TCV
SS
7.6
0.019
1.4
0.049
4.3
7.85
0.206
19.5
8.1
1.876
186.7
4.594
337.5
8.694
652.0
10.908
823.6
10.304
795.4
164
205
-------
APPENDIX D
CHEMOSTAT OPERATION DATA
Chemostat Number 1; Chemostat Volume 1,040 nu
Time
Resi dence
Time
S0
P0„
pH
Absorbance
Time
Resi dence
Time
S0
°P0t
pH
Absorbance
day
days
ug/i-P
Uni ts
OD/inch
day
days
ug/l-P
Uni ts
OD/inch
1
CD
-
-
28
0.
951
64.8
7.0
0.045
2
CO
-
-
29
0.
951
64.8
7.0
0.040
3
oo
-
-
30
0.
951
64.8
6.9
0.037
4
1.39
100
7.9
0.136
31a
0.951
64.8
6.9
0.040
5
1.39
100
8.2
0.120
32
0.
902
64.8
7.3
0.037
6
1.39
100
8.6
0.087
33
0.
902
64.8
7.6
0.040
7
1.39
100
9.3
0.078
34 a
0.
902
64.8
7.5
0.040
8a
1.39
100
7.4
0.065
35
0.756
66.7
-
-
9
1.37
100
-
-
36
0. 756
66.7
6.9
0.040
10
1.37
100
-
-
37
0.
756
66.7
6.9
0.038
11
1.37
100
7.7
0.073
38
0.
756
66.7
6.9
0.036
12
1.37
63
-
-
39
0.
756
66.7
6.9
0.036
13
1.37
63
-
-
40
4.
89
65.1
_
-
14
1.37
63
7.9
0.036
41
4.
89
65.1
8.3
0.096
15
1.37
63
8.0
0.030
42
4.
89
65.1
8.4
0.112
16*
1.37
63
8.0
0.040
43
4.
89
65.1
8.4
0.107
17
3.04
67.5
8.7
0.050
44
4.
89
65.1
8.0
0.106
18
3.04
67.5
8.7
0.053
45a
4.89
65.1
8.3
0.100
19
3.04
67.5
-
-
46
4.
78
65.1
8.0
0.096
20
3.04
67.5
7.9
0.053
47a
4.
78
65.1
8.3
0.094
21
3.04
67.5
7.0
0.048
48
Non-
flow
65.1
_
_
22
2.91
67.5
7. 3
0.045
49
system
65.1
-
-
23
2.91
67.5
7.3
0.050
50
65.1
-
-
24
2.91
67.5
7.2
0.050
51
65.1
8.0
0.120
25
2.91
67.5
7.2
0.055
52
65.1
7.9
0.125
26a
2.91
67.5
7.7
0.050
53a
65.1
8.0
0.120
27
0.951
64.8
~
~
54
55a
65.1
65.1
8.3
8.6
0.125
0.120
Chemostat
Number 2; Chemostat Volume 1,040 me.
1
CO
-
_
28
0.940
63.3
7.0
0.045
2
CO
-
-
29
0.940
63.3
7.0
0.040
3
00
-
-
30
0.940
63.3
7.0
0.030
4
1.39
100
8.0
0.147
31a
0.
940
63.3
7.0
0.037
5
1.39
100
8.2
0.120
32
0.
880
63.3
7.5
0.037
6
1.39
100
8.1
0.080
33
0.880
63.3
7.6
-.042
7
1.39
100
8.3
0.070
34 a
0.880,
63.3
7.6
0.040
8a
1.39
100
7.9
0.065
35
0.
527r
66.7
-
-
9
1.32
100
_
-
36
0.527°
66.7
6.8
0.016
10
1.32
100
-
-
37
0.527^
66.7
7.3
0.010
11
1. 32
100
8.0
0.064
38
0.527
66.7
6.8
0.006
12
1.32
62
-
-
39
0.527b
66.7
6.8
0.005
13
1.32
62
-
-
40
4.69
69.0
-
-
14
1.32
62
7.9
0.040
41
4.69
69.0
8.3
0.054
15
1.32
62
7.8
0.036
42
4,69
69.0
8.4
0.112
16a
1.32
62
8.0
0.040
43
4.
69
69.0
8.4
0.114
17
2.87
68.7
8.4
0.052
44
4.
69
69.0
7.9
0.113
18
2.87
68.7
8.6
0.055
45a
4.
69
69.0
8.2
0.104
19
2.87
68.7
-
-
46
4.
78
69.0
8.0
0.097
20
2.87
>8.7
7.8
0.057
47a
4.
78
69.0
8.2
0.095
21a
2.87
68.7
7.4
0.052
48
Non-
flow
69.0
-
-
22
2.86
68.7
7.5
0.045
49
sys
tern
69.0
-
-
23
2.86
68.7
7.3
0.055
50
69.0
-
-
24
2.86
68.7
7.1
0.057
51
69.0
7.9
0.130
25
2.86
68.7
7.3
0.051
52
69.0
8.0
0.130
26a
2.86
68.7
7.3
0.051
53a
69.0
8.2
0.125
27
0.940
63.3
-
-
54
69.0
8.3
0.125
55
69.0
8.6
0.120
aSampling days
b"Wash-out" occurs.
206
-------
CHEMOSTAT OPERATION DATA (Continued)
Chemostat Number 3; Chemostat Volume 1,020 mj,
Time
Residence
Time
S°po
4
dH
Absorbance
Time
Residence
Time
S°P0k
pH
Absorbance
day
days
ng/Ji-P
Units
OD/inch
day
days
pg/i-P
Uni ts
OD/inch
1
-
-
28
0.823
63.3
7.0
0.040
2
-
-
29
0.823
63.3
7.1
0.030
3
00
-
-
30
0.
823
63.3
7.0
0.030
4
1.78
103
8.0
0.135
31a
0.823
63.3
7.0
0.040
5
1.78
103
8.1
0.109
32
0.773
63.3
7.6
0.040
6
1.78
103
8.3
0.080
33
0.
773
63.3
7.6
0.045
7
1.78
103
8.3
0.065
34a
0.
773
63.3
7.6
0.042
8a
1.78
103
7.9
0.060
35
0.737
66.7
-
-
9
1.13
103
-
-
36
0.
737
66.7
6.8
0.037
10
1.13
103
-
-
37
0.737
66.7
7.7
0.036
11
1.13
64
8.1
0.085
38
0.737
66.7
6.8
0.040
12
1.13
64
-
-
39
0.
737
66.7
6.8
0.038
13
1.13
64
-
-
40
4.
80
66.3
-
-
14
1.13
64
8.0
0.030
41
4.
80
66.3
8.3
0.090
15
1.13
64
7.8
0.030
42
4.
80
66.3
8.3
0.100
1 6a
1.13
64
7.9
0.036
43
4.
80
66.3
8.4
0.114
17
2.61
69.3
8.3
0.043
44
4.
80
66.3
8.0
0.100
18
2.61
69.3
8.6
0.053
45a
4.
80
66.3
7.9
0.100
19
2.61
69.3
-
-
46
4.
69
66.3
7.8
0.096
20
2.61
69.3
7.8
0.041
47a
4.
69
66.3
8.2
0.093
21a
2.61
69.3
7.2
0.037
48
Non-
flow
66.3
-
-
22
2.64
69.3
7.5
0.035
49
sys
tem
66.3
-
-
23
2.64
69,3
7.3
0.043
50
66.3
-
-
24
2.64
69.3
7.0
0.047
51
66.3
7.3
0.125
25
2.64
69.3
7.3
0.055
52
66.3
7.1
0.120
26a
2.64
69.3
7.8
0.053
53a
66.3
8.0
0.120
27
0.823
63.3
-
54
55a
66.3
66.3
8.3
8.4
0.125
0.120
Chemostat
Number 4; Chemostat
Volume 1,
070 mi
1
oo
_
_
28
0.
955
100
7.0
0.065
2
oo
-
.
29
0.
955
100
7.1
0.050
3
oo
-
30
0.
955
100
7.1
0.042
4
1.37
116
8.1
0.203
31 a
0.955
100
7.1
0.050
5
1.37
116
8.3
0.185
32
0.905
100
7.6
0.057
6
1.37
116
8.3
0.110
33
0.
905
100
7.7
0.062
7
1.37
116
8.0
0.083
34a
0.905
100
7.7
0.060
8a
1.37
116
7.9
0.075
35
0.
556
100
-
-
9
1.29
116
-
-
36
0.
556
100
6.9
0.060
10
1.29
116
-
-
37
0.
556
100
7.7
0.057
11
1.29
116
8.1
0.065
38
0.
556
100
7.0
0.060
12
1.29
99
-
-
39
0.
556
100
6.9
0.060
13
1.29
99
-
-
40
4.
82
103
-
-
14
1.29
99
8.2
0.045
41
4.
82
103
8.3
0.140
15
1.29
99
7.9
0.045
42
4.
82
103
8.4
0.190
16a
1.29
99
7.9
0.050
43
4.
82
103
8.4
0.185
17
3.06
104
8.3
0.067
44
4.
82
103
7.6
0.183
18
3.06
104
8.4
0.077
45a
4.
82
103
7.9
0.180
19
3.06
104
-
-
46
4.
82
103
7.5
0.180
20
3.06
104
7.9
0.080
47a
4.
82
103
8.0
0.175
21a
3.06
104
7.2
0.078
48
Non-
flow
103
-
-
22
2.87
104
7.5
0.060
49
sys
tem
103
-
-
23
2.87
104
7.3
0.065
50
103
-
-
24
2.87
104
7.0
0.067
51
103
7.1
0.210
25
2.87
104
7.4
0.080
52
103
6.9
0.213
26a
2.87
104
7.8
0.080
53a
103
7.5
0.215
27
0.955
100
54
55a
103
103
8.3
8.5
0.220
0.220
aSampling days
207
-------
CHEMOSTAT OPERATION DATA (Continued)
Chemostat Number 5; Chemostat Volume 1,070 rnn
Time
Resi dence
Time
S°P0lt
pH
Absorbance
Time
Residence
Time
s°poIt
pH
Absorbance
day
days
ug/2-P
Uni ts
OD/inch
day
days
ug/n-P
Units
OD/inch
1
00
_
28
0.967
93.3
7.1
0.060
2
00
-
-
29
0.
967
93.3
7.2
0.047
3
00
-
-
30
0.
967
93.3
7.1
0.045
4
1.43
120
8.0
0.165
31a
0.
967
93. 3
7.2
0.052
5
1.43
120
8.3
0.141
32
0.
861
93.3
7.6
0.050
6
1.43
120
8.2
0.100
33
0.
861
93.3
7.7
0.052
7
1.43
120
8.3
0.085
34 a
0.
861
93.3
7.7
0.053
8^
1.43
120
8.0
0.081
35
0.
740
100
-
-
9
1.36
120
-
-
36
0.740
100
6.9
0.056
10
1.36
120
-
-
37
0.
740
100
7.7
0.058
n
1.36
120
8.0
0.075
38
0
740
100
6.8
0.055
12
1.36
99
-
-
39
0.
740
100
6.9
0.058
13
1.36
99
-
-
40
5.
10
106
-
-
14
1.36
99
8.3
0.063
41
5.
10
106
8.3
0.162
15
1.36
99
7.9
0.064
42
5.
10
106
8.4
0.194
16a
1.36
99
7.8
0.060
43
5.
10
106
8.4
0.198
17
3.06
104
8.6
0.085
44
5.
10
106
7.7
0.183
18
3.06
104
8.2
0.088
45a
5.
10
106
8.0
0.184
19
3.06
104
-
-
46
5.
03
106
7.4
0.175
20
3.06
104
7.7
0.085
47a
5.03
106
8.1
0.180
21a
3.06
104
7.2
0.088
48
Non-
flow
106
-
-
22
2.86
104
7.5
0.070
49
system
106
-
-
23
2.86
104
7.3
0.077
50
106
-
-
24
2.86
104
7.4
0.076
51
106
7.2
0.220
25
2.86
104
7.4
0.085
52
106
7.1
0.220
26a
2.86
104
7.8
0.080
53a
106
7.6
0.215
27
0.967
93.3
•
"
54
55a
106
106
8.3
8.3
0.220
0.210
Chemostat
Number 6; Chemostat Volume 1,080 mi
1
00
-
28
0.
910
96.3
7.1
0.056
2
00
.
.
29
0.
910
96.3
7.3
0.045
3
00
-
-
30
0.
910
96.3
7.2
0.048
4
1.38
117
8.1
0.135
31a
0.910
96.3
7.2
0.049
5
1.38
117
8.2
0.115
32
0.
869
96.3
7.6
0.054
6
1.38
117
8.3
0.089
33
0.
869
96.3
7.7
0.056
7
1.38
117
8.4
0.085
34a
0.
869
96.3
7.7
0.055
8a
1.38
117
8.2
0.083
35
0.668
100
-
-
9
1.19
117
.
-
36
0.668
TOO
6.8
0.056
10
1.19
117
-
-
37
0.668
100
7.7
0.056
11
1.19
117
8.0
0.066
38
0.
668
100
6.9
0.060
12
1.19
94
-
-
39
0.
668
100
6.9
0.057
13
1.19
94
.
40
5.08
101
-
-
14
1.19
94
8.3
0.050
41
5.08
101
8.3
0.140
15
1.19
94
8.0
0.053
42
5.
08
101
8.3
0.194
16a
1.19
94
7.9
0.060
43
5.08
101
8.4
0.195
17
2.69
100
8.0
0.085
44
5.
08
101
7.5
0.189
18
2.69
100
7.9
0.095
45 a
5.
08
101
8.0
0.190
19
2.69
100
_
46
4.97
101
7.4
0.180
20
2.69
100
7.7
0.085
47a
4.
97
101
8.0
0.180
21a
2.69
100
7.2
0.088
48
Non-
flow
101
-
-
22
2.74
100
7.5
0.065
49
system
101
-
-
23
2.74
100
7.3
0.070
50
101
-
-
24
2.74
100
7.0
0.073
51
101
7.0
0.235
25
2.74
100
7.4
0.085
52
101
7.0
0.230
26a
2.74
100
7.9
0.075
53a
101
7.9
0.230
27
0.910
96.3
_
-
54
101
8.3
0.230
55a
101
8.5
0.235
Sampling days.
208
-------
CHEMOSTAT OPERATION DATA (Continued)
Chemostat Number 7; Chemostat Volume 1,050 im
Time
Residence
Time
S°POl,
pH
Absorbance
T ime
Resi dence
Time
SoP0u
PH
Absorbance
day
days
ug/i-P
Units
OD/inch
day
days
ug/n-P
Units
OD/inch
1
oo
_
.
28
0.960
128
7.2
0.085
2
CO
-
-
29
0.
960
128
7.3
0.075
3
00
-
-
30
0.960
128
7.2
0.070
4
1.48
136
8.3
0.165
31a
0.960
128
7.2
0.080
5
1.48
136
8.2
0.148
32
0.
906
128
7.6
0.072
6
1.48
136
8.3
0.118
33
0.906
128
7.9
0.070
7
1.48
136
8.6
0.110
34 a
0.
906
128
7.9
0.075
8a
1.48
136
8.4
0.106
35
0.
726
138
-
-
9
1.27
136
-
-
36
0.
726
138
7.0
0.069
10
1.27
136
-
-
37
0.
726
138
7.8
0.070
11
1.27
136
8.2
0.095
38
0.
726
138
7.0
0.073
12
1.27
129
-
-
39
0.
726
138
7.0
0.075
13
1.27
129
-
-
40
5.
08
137
-
-
14
1.27
129
8.3
0.066
41
5.
08
137
8.3
0.195
15
1.27
129
7.9
0.070
42
5.
08
137
8.4
0.258
16a
1.27
129
7.9
0.077
43
5.
08
137
8.4
0.256
17
2.74
135
8.1
0.110
44
5.
08
137
7.6
0.240
18
2.74
135
8.0
0.130
45a
5.08
137
8.0
0.247
19
2.74
135
-
-
46
4.93
137
7.4
0.233
20
2.74
135
7.6
0.130
47a
4.
93
137
8.0
0.223
21a
2.74
135
7.3
0.125
48
Nnn-
flnw
137
-
-
22
2.88
135
7.5
0.116
49
svstem
137
-
-
23
2.88
135
7.3
0.127
50
137
-
-
24
2.88
135
7.1
0.120
51
137
7.0
0.300
25
2.88
135
7.5
0.140
52
137
7.0
0.300
26a
2.88
135
7.9
0.135
53a
137
7.9
0.290
27
0.960
128
-
-
54
55a
137
137
8.3
8.5
0.295
0.286
Chemostat Number 8; Chemostat Volume 1,030 mn
1
00
_
_
28
0.
873
130
7.3
0.085
2
00
_
_
29
0.
873
130
7.3
0.075
3
00
-
-
30
0.
873
130
7.3
0.080
4
1.45
136
8.2
0.165
31a
0.873
130
7.3
0.085
5
1.45
136
8.2
0.140
32
0.
876
130
7.7
0.075
6
1.45
136
8.4
0.110
33
0.876
130
7.9
0.075
7
1.45
136
8.7
0.099
34 a
0.876
130
8.0
0.078
8a
1.45
136
8.4
0.090
35
0.
655
138
-
-
9
1.20
136
-
-
36
0.
655
138
7.0
0.070
10
1.20
136
_
_
37
0.
655
138
7.9
0.065
11
1.20
136
8.1
0.075
38
0.
655
138
7.2
0.067
12
1.20
130
-
-
39
0.
655
138
7.0
0.070
13
1.20
130
_
-
40
4.
58
137
-
-
14
1.20
130
8.3
0.083
41
4.
58
137
8.3
0.170
15
1.20
130
8.0
0.083
42
4.
58
137
8.5
0.220
16a
1.20
130
8.0
0.080
43
4.
58
137
8.4
0.230
17
2.69
136
8.8
0.126
44
4.
58
137
7.5
0.225
18
2.69
136
7.8
0.147
45a
4.
58
137
7.9
0.228
19
2.69
136
-
-
46
4.
55
137
7.4
0.225
20
2.69
136
7.6
0.140
47a
4.
55
137
8.0
0.223
2i a
2.69
136
7.4
0.140
48
Non-
flow
137
-
-
22
2.68
136
7.5
0.127
49
system
137
-
-
23
2.68
136
7.3
0.120
50
137
-
-
24
2.68
136
7.1
0.126
51
137
7.2
0.280
25
2.68
136
7.5
0.135
52
137
7.5
0.275
26a
2.68
136
8.3
0.130
53a
137
7.9
0.280
27
0.873
130
-
-
54
137
8.3
0.280
55a
T
137
8.5
0.280
aSamplinq days.
209
-------
CHEMOSTAT OPERATION DATA (Continued)
Chemostat Number 9; Chemostat Volume 1 ,055 rm.
Time
Residence
Time
s°po1|
pH
Absorbance
Time
Residence
Time
S°po„
pH
Absorbance
day
days
ng/8-
P Units
OD/inch
day
days
wg/£-P
Uni ts
OD/inch
1
CO
_
28
0.
870
128
7.3
0.087
2
CO
-
-
29
0.
870
128
7.3
0.075
3
«
-
-
30
0.
870
128
7.3
0.075
4
1.30
137
8.1
0.073
31a
0.
870
128
7.3
0.078
5
1.30
137
8.4
0.163
32
0.
829
128
7.7
0.070
6
1.30
137
8.3
0.093
33
0.829
128
8.0
0.065
7
1.30
137
8.5
0.099
34a
0.
829
128
8.0
0.070
8a
1.30
137
8.4
0.095
35
0.739
138
-
-
9
1.25
137
-
-
36
0.739
138
6.9
0.057
10
1.25
137
-
-
37
0.
739
138
7.8
0.061
11
1.25
137
8.1
0.065
38
0.739
138
6.9
0.068
12
1.25
129
-
-
39
0.
739
138
7.0
0.065
13
1.25
129
-
-
40
4.
89
136
-
-
14
1.25
129
8.3
0.081
41
4.
89
136
8.3
0.170
15
1.25
129
8.0
0.076
42
4.
89
136
8.3
0.230
16a
1.25
129
7.9
0.070
43
4.
89
136
8.4
0.250
17
2.68
133
8.7
0.105
44
4.
89
136
7.5
0.245
18
2.68
133
7.8
0.127
45a
4.
89
136
8.0
0.240
19
2.68
133
-
-
46
4.
85
136
7.4
0.233
20
2.68
133
7.5
0.135
47a
4.
85
136
8.0
0.235
21a
2.68
133
7.3
0.135
48
Non-
flow
136
-
-
22
2.63
133
7.5
0.117
49
system
136
-
-
23
2.63
133
7.3
0.117
50
136
-
-
24
2.63
133
7.1
0.123
51
136
7.4
0.285
25
2.63
133
7.5
0.136
52
136
7.2
0.290
26a
2.63
133
8.0
0.140
53a
136
7.8
0.290
27
0.870
128
-
54
55a
136
136
8.3
8.5
0.285
0.290
Chemostat
Number 10; Chemostat
Volume 1,045
mi
1
CO
_
28
0.861
161
7.3
0.105
2
oo
-
-
29
0.861
161
7.3
0.096
3
00
-
-
30
0.
861
161
7.4
0.097
4
1.32
170
8.1
0.110
31a
0.861
161
7.4
0.098
5
1.32
170
8.5
0.135
32
0.
817
161
7.8
0.088
6
1.32
170
8.5
0.125
33
0.
817
161
8.0
0.084
7
1.32
170
8.5
0.135
34a
0.817
161
8.0
0.085
8a
1.32
170
8.5
0.135
35
0.
718
175
-
-
9
1.22
170
-
-
36
0.
718
175
7.0
0.075
10
1.22
170
-
-
37
0.
718
175
7.8
0.070
11
1.22
170
8.8
0.130
38
0.
718
175
7.0
0.068
12
1.22
161
-
-
39
0.718
175
7.0
0.070
13
1.22
161
-
-
40
4.
78
172
-
-
14
1.22
161
8.5
0.103
41
4.
78
172
8.3
0.170
15
1.22
161
8.1
0.106
42
4.
78
172
8.5
0.260
16a
1.22
161
8.1
0.110
43
4.
78
172
8.5
0.282
17
2.62
166
8.4
0.150
44
4.
78
172
7.5
0.301
18
2.62
166
7.8
0.175
45a
4.
78
172
8.0
0.300
19
2.62
166
-
-
46
4.
62
172
7.4
0.293
20
2.62
166
7.6
0.180
47a
4.62
172
8.0
0.285
21a
2.62
166
7.2
0.183
48
Non-
flow
172
-
-
22
2.61
166
7.5
0.157
49
system
172
-
-
23
2.61
166
7.3
0.160
50
172
-
-
24
2.61
166
7.2
0.165
51
172
8.2
0.355
25
2.61
166
7.5
0.190
52
172
8.1
0.360
26a
2.61
166
8.1
0.190
53a
172
8.4
0.360
27
0.861
161
_
-
54
172
8.3
0.365
55a
172
8.5
0.365
aSamp1ing days.
210
-------
CHEMOSTAT OPERATION DATA (Continued)
Chemostat Number 11; Chemostat Volume 1,055 mi
Time
Resi dence
Time
s°pou
PH
Absorbance
Time
Resi dence
Time
So,
pH
Absorbance
day
days
ug/i-P
Units
OD/i nch
day
days
ug/a-P
Units
OD/i nch
1
no
_
-
28
0.
924
162
7.3
0.115
2
»
-
-
29
0.924
162
7.4
0.100
3
CO
-
-
30
0.924
162
7.4
0.098
4
1.35
171
8.9
0.120
31 a
0.
924
162
7.4
0.100
5
1.35
171
8.8
0.150
32
0.
865
162
7.4
0.096
6
1.35
171
9.0
0.140
33
0.
865
162
8.2
0.097
7
1.35
171
9.2
0.135
34a
0.
865
162
8.2
0.098
8a
1.35
171
8.7
0.120
35
0.
763
175
-
-
9
1.22
171
-
-
36
0.763
175
7.0
0.086
10
1.22
171
-
-
37
0.
763
175
7.8
0.077
11
1.22
171
8.7
0.123
38
0.763
175
6.9
0.080
12
1.22
163
-
-
39
0.763
175
7.0
0.082
13
1.22
163
-
-
40
5.
18
172
-
-
14
1.22
163
8.8
0.112
41
5.
18
172
8.3
0.180
15
1.22
163
8.3
0.113
42
5.
18
172
8.6
0.240
16a
1.22
163
8.0
0.113
43
5.
18
172
8.5
0.255
17
2.68
172
8.3
0.153
44
5.
18
172
7.5
0.270
18
2.68
172
7.8
0.175
45a
5.
18
172
8.1
0.272
19
2.68
172
-
-
46
5.
18
172
7.4
0.270
20
2.68
172
7.6
0.190
47a
5.
18
172
8.0
0.280
21a
2.68
172
7.3
0.185
48
Non-
flow
172
-
-
22
2.58
172
7.5
0.175
49
sys tem
172
-
-
23
2.58
172
7.3
0.170
50
172
-
-
24
2.58
172
7.2
0.173
51
172
8.0
0.335
25
2.58
172
7.5
0.195
52a
172
7.8
0.330
26a
2.58
172
8.0
0.190
53
172
8.0
0.335
27
0.924
162
-
54
55
172
172
8.3
8.5
0.330
0.335
Chemostat Number 12; Chemostat Volume 1,055 mi
1
ao
.
.
28
0.920
164
7.3
0.120
2
oo
_
29
0.920
164
7.4
0.100
3
co
_
-
30
0.920
164
7.4
0.100
4
1.45
175
8.6
0.183
31a
0.920
164
7.4
0.105
5
1.45
175
8.4
0.185
32
0.866
164
7.8
0.100
6
1.45
175
9.2
0.145
33
0.866
164
8.2
0.095
7
1.45
175
9.3
0.125
34 a
0.866
164
8.2
0.099
8a
1.45
175
8.8
0.115
35
0.785
175
-
-
9
1.32
175
.
36
0.785
175
7.1
0.095
10
1.32
175
_
37
0.785
175
7.7
0.085
11
1.32
175
8.2
0.100
38
0.785
175
7.2
0.086
12
1.32
165
-
-
39
0.785
175
7.1
0.085
13
1.32
165
_
.
40
5.51
171
-
-
14
1.32
165
8.6
0.090
41
5.51
171
8.3
0.205
15
1.32
165
8.3
0.098
42
5.51
171
8.5
0.280
16a
1.32
165
8.0
0.105
43
5.51
171
8.5
0.280
17
2.86
169
8.2
0.149
44
5.51
171
7.4
0.285
18
2.86
169
7.8
0.172
45a
5.51
171
7.9
0.280
19
2.86
169
-
-
46
5.42
171
7.4
0.275
20
2.86
169
7.5
0.163
47a
5.42
171
7.9
0.278
21a
2.86
169
7.3
0.162
48
Non-
flow
171
-
-
22
2.82
169
7.5
0.165
49
system
171
-
-
23
2.82
169
7.3
0.170
50
171
-
-
24
2.82
169
7.2
0.172
51
171
7.8
0.330
25
2.82
169
7.5
0.185
52
171
7.7
0.330
26a
2.82
169
8.0
0.175
53a
171
8.4
0.340
27
0.920
164
-
-
54
171
8.4
0.345
55
171
8.5
0.345
aSamplinq days.
211
------- |